Evaluation of Agolin® Pig on sow and piglet performance and sow fecal microbial communities during lactation

Abstract

Essential oils are increasingly evaluated as alternatives to antibiotics in swine nutrition. This pilot study investigated the effects of Agolin Pig, a blend of microencapsulated essential oils, on sow performance, piglet growth, and sow fecal microbiota during lactation. Twenty-five sows were randomly assigned to either a control group (CON; n = 13) or a treatment group supplemented with Agolin Pig (AGO; n = 12; 200 ppm) from two weeks pre-farrowing through lactation. Diets for both groups met NRC (2012) nutrient requirements. Sow average daily feed intake (ADFI), body weight, body condition score (BCS), and Knauer Caliper measurements were collected at baseline, one week post-farrowing, and weaning. Piglet average daily gain (ADG) was recorded throughout lactation. Sow fecal samples were collected on d-14, d-7, d0, d8, and d16 relative to farrowing for microbiome analysis. Performance data were analyzed using PROC GLM or MIXED (SAS 9.4), with parity, number nursed, number weaned, and lactation week included as covariates. Microbiota data were analyzed in QIIME2 (v2022.8) and R (v4.2.3). Agolin Pig supplementation did not affect sow ADFI, piglet ADG, or colostrum and milk composition. However, AGO sows lost significantly less body condition during lactation, as measured by the Knauer Caliper (P = 0.010). Beta diversity differed between treatments on d8 (P = 0.029) and tended to differ on d16 (P = 0.066). Additionally, Clostridium and Streptococcus increased in CON but not AGO sows on d8 and d16, respectively. In summary, Agolin Pig supplementation reduced body condition loss and altered sow fecal microbiota diversity during lactation. These findings highlight the potential of essential oils to support sow health and productivity and warrant confirmation in larger studies.

This pilot study evaluated the potential of Agolin Pig, a novel blend of microencapsulated essential oils, as a natural alternative to antibiotics in swine diets, demonstrating a potential ability to maintain sow body condition and promote alterations in sow fecal microbiota diversity during lactation.

Introduction

Modern sows have been genetically selected for high prolificacy resulting in larger litter sizes and piglets with greater lean tissue deposition (Ward et al. 2020). Lactation in sows is marked by significant changes in body weight and composition due to the substantial energy and amino acid demands of milk production which have only increased with the increase in litter size (Farmer 2022). Consequently, modern sows require an even greater amount of nutritional intake to meet the demands of increased milk production with voluntary feed intake typically being insufficient, culminating in a negative nutrient balance and mobilization of the sow’s body reserves during lactation (Silva et al. 2021). Additionally, the increase in litter size is accompanied by reduced average birth weights, increased vulnerability, and limited growth potential of the light birthweight pigs (Moreira et al. 2020). To address these challenges, feeding strategies for reproductive females have focused on increasing feed intake in lactation to sustain body fat reserves and support colostrum production, enhancing farrowing and lactation performance, and maintaining body composition to optimize subsequent reproductive performance (Theil et al. 2023). Furthermore, the health of piglets is directly related to sow health via antibodies in colostrum and through the initial colonization of their gastrointestinal (GI) tract by the gut microbiota of the sow and the environment (Lim et al. 2023).

Previously, antibiotics were commonly used in sows to reduce bacterial infections, improve lactation performance, and prevent the spread of infections to susceptible piglets (Kyriakis et al. 1996). However, growing concerns over the use of antibiotics, along with bans placed in the European Union (EU), have led to the exploration of numerous alternatives. There is an increasing need to find natural alternatives to antibiotics to aid in animal health and assist in animal growth performance. Natural compounds, such as phytogenics, have been explored to increase feed intake during lactation, improve the intestinal health of sows, and improve litter health and performance (Zhaikai Zeng et al. 2015; Nowland et al. 2021; Silva et al. 2021; Santos et al. 2023). Phytogenics are plant-derived natural bioactive compounds made of volatile and lipophilic substances containing active compounds such as carvacrol, thymol, cinnamaldehyde, limonene, and anethole (Camargo et al. 2023). Among these, essential oils (EOs) have shown potential to improve sow daily feed intake during lactation, increase litter weight gain, reduce diarrhea in suckling pigs, and alter the gut microbiome using in vitro models (Allan and Bilkei 2005; Leyva-Lopez et al. 2017; Hall et al. 2021; Wang et al. 2022; Santos et al. 2023).

Previous research has explored the potential of EOs to improve gut health in sows. Oregano EO (OEO) contains the active ingredients of thymol and carvacrol, reported to have antimicrobial and antioxidative properties (Lambert et al. 2001). Piglets of sows supplemented with OEO have demonstrated improved growth performance and reduced incidence of therapeutic treatments (Park et al. 2016; Hall et al. 2021). The piglets of sows that received a blend of anethole, cinnamaldehyde, and eugenol EOs have demonstrated improved feed intake and growth post-weaning compared to piglets from sows that received a basal diet in lactation (Blavi et al. 2016). Sows supplemented with OEO during lactation have shown an increase in the relative abundance of Lactobacillaceae family (Hall et al. 2021; Zhang et al. 2024) and in bacterial families known to be important in fiber digestion (Fibrobacteriaceae and Akkermansiaceae; Hall et al. 2021). Blends of EO appear to improve average daily gain (ADG) and average daily feed intake (ADFI) compared to individual supplementation of EOs in weaned pigs (Zhaikai Zeng et al. 2015). While several studies explored EO supplementation in weaned pigs, there is limited research exploring EO-blends supplementation in sows during lactation and the impacts on the sows’ fecal microbiota.

In this study, we hypothesized that supplementing sows with a blend of EOs in the feed at the end of gestation and throughout lactation would enhance daily feed intake in sows during the lactation period, enhance litter performance, and reduce body weight and condition loss by the end of lactation, with modulatory effects on sow fecal microbiome. This preliminary study aimed to evaluate the effects of Agolin Pig, a microencapsulated EO blend, on the performance parameters of lactating sows and their suckling piglets, as well as on the sow fecal microbiota.

Materials and methods

Animals and diets

All procedures were approved by the Purdue University Animal Care and Use Committee (#1909001949). A total of 25 sows (Landrace x Large White or Landrace x York) and their offspring [Dam genetics x Duroc] were used in this study. Sows were blocked by parity and randomly assigned to either the control (CON, n = 13) or treatment (AGO, n = 12) groups. Initial body weights (BW) and body condition scores (BCS) were balanced across treatment groups. Both groups were fed a standard basal diet that met or exceeded NRC standards (NRC 2012) with the AGO diet including an in-house premix containing 200 ppm of Agolin Pig (a blend of microencapsulated EOs, including clove, eugenol, thymol, piperine, and geraniol; Alltech, Nicholasville, KY). Upon entry into the farrowing house, approximately two weeks prior to farrowing, CON received a basal gestation diet with no additional supplementation, and AGO received a basal gestation diet with 200 ppm of Agolin Pig. Both groups were transitioned to a basal lactation diet on their day of farrowing but remained on their respective treatments within the lactation diet (Table 1). Sows ranged in parity from gilts (parity 1) to parity 6, with parities similarly distributed between the two treatment groups, with an average parity of 3.20 for CON and 3.04 for AGO. Two cc of prostaglandin (Lutalyse) followed 6 h later by 1 cc of oxytocin (20 IU/mL) were administered to sows that had not farrowed by day 114 of gestation to induce farrowing on the subsequent day.

Table 1.

Diet compositions and chemical analysis (as-fed basis).

	CONTROL (CON)		AGOLIN (AGO)
Ingredient, %	Gestation1	Lactation2	Gestation3	Lactation2
Corn	53.92	46.09	53.92	46.09
SBM, 47.5% CP	11.57	32.00	11.57	32.00
DDGS, 7% fat	30.00	15.00	30.00	15.00
Swine Grease	1.00	3.00	1.00	3.00
Limestone	1.65	1.58	1.65	1.58
MonoCal. Phos.	0.00	0.70	0.00	0.70
Sow Vit/TM premix4	0.15	0.15	0.15	0.15
Choline Cloride (60%)	0.10	0.10	0.10	0.10
Carnitine/Chromium premix	0.01	0.01	0.01	0.01
Phytase5	0.15	0.15	0.15	0.15
Salt	0.50	0.50	0.50	0.50
Larvacide6	0.33	0.10	0.33	0.10
Mycotoxin Prev.7	0.25	0.25	0.25	0.25
Sow trace mineral supplement8	0.08	0.08	0.08	0.08
Kemgest9	0.10	0.10	0.10	0.10
Control Premix10	0.20	0.20	—	—
Agolin-Pig Premix11	—	—	0.20	0.20
Total	100.00	100.00	100.00	100.00
Calculated Analysis12
ME, Kcal/kg	3296.90	3363.70	3296.90	3363.70
NE, Kcal/kg	2437.20	2456.20	2437.20	2456.20
CP, %	17.87	23.03	17.87	23.03
Total Lysine, %	0.75	1.20	0.75	1.20
SID Lys, %	0.55	1.00	0.55	1.00
Ca, %	0.70	0.85	0.70	0.85
P, %	0.46	0.62	0.46	0.62
Avail. P., %	0.36	0.45	0.36	0.45
Analyzed Analysis13
CP14, %	17.43	22.49	17.38	23.09
Total Lysine, %	0.74	1.21	0.80	1.37
Moisture, %	12.62	10.92	12.34	11.20
Crude Fat, %	4.98	5.74	5.09	5.39
Crude Fiber, %	3.41	2.76	3.81	2.95
Ash, %	4.88	5.60	4.58	5.51
Ca, %	0.83	0.97	0.76	0.85
P, %	0.44	0.54	0.43	0.55

¹All sows were fed control gestation diet from d 35 post-breeding until approximately 2 weeks prior to d 112 of gestation. Sows in the CON group received control gestation diet upon entry into the farrowing house until farrowing (approximately 2 weeks).

²Diet was fed from the day of farrowing until weaning.

³Sows in the AGO group received Agolin-supplemented gestation diet upon entry into the farrowing house until farrowing (approximately 2 weeks).

⁴Provided per kg of diet: vitamin A, 9,010 IU; vitamin D3, 2,245 IU; vitamin E, 66 IU; vitamin K, 2.21 mg; riboflavin, 7.1 mg; pantothenic acid, 20.1 mg; niacin, 39.9 mg; B12, 37 µg; biotin, 0.24 mg; folic acid, 1.73 mg; thiamine, 2.21 mg; pyrdoxine-B6, 4.00 mg; iron, 100 mg; zinc, 120 mg; manganese, 50 mg; copper, 20 mg; iodine, 0.70 mg; selenium, 0.3 mg.

⁵Enzae (Cargill Animal Nutrition, Minneapolis, MN), provided 1,500 phytase units/kg.

⁶ClariFly Larvicide (Central Life Sciences, Dallas, TX), a feed-through larvicide to prevent the development of flies in manure.

⁷Defusion Plus (Cargill Animal Nutrition, Minneapolis, MN) is a mixture of preservatives and other ingredients intended for use in swine rations to bind mycotoxins.

⁸Availa Sow (Zinpro Corporation, Eden Prairie, MN; AAFCO No. 57.150). Minerals provided per kg of diet: zinc, 6.67%; manganese, 2.67%; and copper, 1.34%. Proximates: moisture, 1.0–2.5%; ash, 15.6%; crude protein, 25.5%; crude fiber, 14.1%.

⁹KEM-GEST (Kemin Industries, Des Moines, IA) is a blend of organic and inorganic acids to acidify swine feed.

¹⁰Control premix included finely ground corn included at the same amount as the test premix to reduce variability in feed mixing.

¹¹Agolin Pig (Alltech, Nicholasville, KY) is a microencapsulated blend of essential oils (clove, eugenol, thymol, piperine, and geraniol) included at 200 ppm blended with finely ground corn as a premix.

¹²Calculated nutrients were targeted to meet or exceed the NRC (2012).

¹³Samples were collected weekly, pooled by diet phase and treatment, ground, and subsampled at Purdue University and assayed by the University of Missouri-Columbia, Agricultural Experiment Station Chemical Laboratories (Columbia, MO).

¹⁴Calculated as N × 6.25.

All animals were limit-fed before farrowing (2.72 kg/d) and were transitioned to ad libitum feeding on the day of farrowing. Each morning, feed remaining in the feeder was vacuumed, weighed, and returned to the feeder (weighback). Weighbacks were used to measure the feed consumed to calculate the average daily feed intake (ADFI) during lactation. Weighbacks were subtracted from the daily feed additions to calculate feed consumption. On d7 post-farrowing and at weaning, body condition score (1 to 5 scale; Coffey et al. 1999) was recorded for each sow by a single evaluator in addition to body weight using a floor scale and Knauer Sow Caliper measurement at the 12^th rib. Individual piglet weights were recorded within 24 h of farrowing, on d7 and 14 post-farrowing, and at weaning. Crossfostering was recorded and occurred within 24–48 h of farrowing within treatment groups.

Milk collection and analysis

Colostrum was collected from all sows within 2 h of the onset of farrowing. Between 50–100 mL was collected from all teats and mixed to create a homogenous, representative sample. All samples were stored at -20°C until analysis.

Milk samples were collected on d3 and 14 post-farrowing. To prepare for collection, piglets were removed from the sow for 45 minutes and sows were injected with 1 mL of oxytocin (20 IU/mL) via the ear vein. Three functioning mammary glands were selected along the midline of the sow, and all milk was completely extracted from each selected teat (Atwood and Hartmann 1992). Samples from each teat were kept separate as well as mixed to create a composite, homogenous sample of 30–50 mL and stored at −20°C until analysis. One sow in the AGO group was removed from milk analysis due to the loss of milk samples in storage.

Total protein concentration

Colostrum and milk samples were analyzed for total protein concentration using bovine serum albumin (BSA) standards using the Pierce Coomassie (Bradford) Protein Assay (Thermo Scientific, Cat. No. 23200) following the manufacturer’s instructions. For colostrum, the composite sample was measured in triplicate and averaged for each sow. For d3 and d14 milk samples, protein concentration from each individual teat was measured and averaged per sow per day for statistical analysis.

Immunoglobulin G (IgG) and A (IgA)

A dilution of 1:1,000,000 of colostrum total protein was used to quantify IgG (mg/mL), which was measured using an IgG ELISA kit (Bethyl Cat. No. E101-104) according to the manufacturer’s protocols. Similarly, a dilution of 1:30,000 of colostrum total protein was used to determine IgA (mg/mL) concentration, which was measured using an IgA ELISA kit (Bethyl Cat. No. E101-102) according to the manufacturer’s protocols.

Creamatocrit (colostrum and milk fat content)

Creamatocrit procedures were conducted to measure and calculate the fat percentage and caloric content of colostrum and milk on d0, 3, and 14. Approximately 75 µL of homogenized milk were loaded into a hematocrit tube and centrifuged at 3,400 × g for 15 minutes. For colostrum samples, each homogenized sample was measured in triplicate and averaged per sow. For d3 and 14 milk samples, a sample from each teat was measured in triplicate, and all samples were averaged. The total length of the sample and the length of the lipid layer at the top were measured digitally using a macros program in JMP (Alexander Pasternak, PhD, Purdue University, West Lafayette, IN). Fat percentage was calculated as the ratio between lipid length to total length. Fat content was calculated according to the formula: Fat (g/L) = ([lipid (%) − 0.59]/0.146; Lucas et al. 1978).

Microbiome analysis

Fresh fecal samples were collected from all sows via rectal stimulation upon entry into the farrowing house before the start of supplementation (d-14), 7 (d-7) days after entry into the farrowing house and having been fed their respective treatments, on the day of farrowing (d0), 8 days post-farrowing (d8), and 16 days post-farrowing (d16). Fecal sample bags were massaged by hand at collection to ensure the sample were homogenous throughout and repeatable. All samples were stored at -20°C until analysis. DNA was extracted from each sample using QIAamp PowerFecal Pro DNA kit (QIAGEN, Germantown, MD, USA) following the manufacturer’s protocol. Extracted DNA was stored at -20°C for 1–4 weeks prior to 16S rRNA gene amplicon sequencing and qPCR.

Amplification of the V4 region of the 16S rRNA gene was conducted using primers 515F (5′-GTGCCAGCMGCCGCGGTAA) and 806R (5’-GGACTACHVHHHTWTCTAAT). PCR amplifications were conducted in 20 µL reaction volumes in 96-well plates which were then normalized (SequalPrep Normalization Plate; Invitrogen) and pooled into a single library. A known mock community (20 Strain Even Mix Genomic Material; ATCC MSA-1002TM) and a control (water) were included along with the pooled libraries. Sequencing was conducted on the Illumina MiSeq platform (2x250 paired-end sequencing) at the Purdue University Genomics Core Facility.

All sequences were imported into QIIME2 (v2022.8) as demultiplexed EMP paired-end sequences for further analysis (Bolyen et al. 2019). Forward and reverse reads were trimmed at 13 bp and truncated where quality significantly declined (forward reads were truncated at 250 bp and reverse reads were truncated at 130 bp). Sequences were merged, denoised, chimeric sequences removed, and amplicon sequence variants (ASVs) were created using the DADA2 paired-end pipeline (Callahan et al. 2016). Of the initial 2,637,257 reads, a total of 2,172,213 reads remained after the quality filtering step. Sequences were rarefied to 10,806 sequences per sample, resulting in a total of 1,525 ASVs for further analyses. Three samples were excluded at this sub-sampling depth (AGO, Sow ID 57 (d-14), 44 (d-14), and 46 (D16)). Sequences were aligned to the GreenGenes database (v13_8 99% OTUs) for taxonomic assignment.

Statistical and visualization analysis

Sow performance data were analyzed using PROC GLM procedure in SAS 9.4 (Cary, North Carolina). The main effects included treatment with parity class (P1 + P2 vs P ≥ 3), which were grouped to ensure balanced distribution of sows for this pilot study, along with the number of piglets nursed, the number of piglets weaned, and lactation length as covariates where appropriate. For sow feed intake, data were analyzed with PROC MIXED for repeated measures of sows with the number nursed used as a covariate where appropriate. For analysis of protein composition in milk at days 0, 3, and 14 of lactation, data were analyzed using PROC MIXED with treatment and parity class as main effects with day of collection as a repeated measure and a random effect of sow. Piglet data were analyzed using PROC GLM procedure in SAS 9.4 (Cary, North Carolina) with treatment as the main effect with parity class and lactation length used as covariates where appropriate. Results are reported as least squares means with standard errors of the mean and were considered statistically significant at P ≤ 0.05 and tendencies noted for 0.05 < P ≤ 0.10.

All statistical analysis and visualizations for microbiome results were conducted in QIIME2 (v2022.8) and R (v4.2.3) using the following R packages: ggplot2, phyloseq, tidyverse, vegan, and qiime2R. Alpha-diversity was measured via Observed Features (richness), Faith’s PD (phylogenetic diversity), Pielou (evenness), Shannon index (richness and evenness), and Chao1 (richness) metrics (Chao 1984; DeSantis et al. 2006). Statistical significance for alpha-diversity indices was assessed using ANOVA (if data were normally distributed) and Kruskal-Wallis (if data were not normally distributed; R v4.2.3). Normality was tested with Shapiro’s test. Beta-diversity was assessed using Bray-Curtis Dissimilarity, Jaccard Distance, Unweighted UniFrac, and Weighted UniFrac. Permutational multivariate analysis of variance (PERMANOVA) test and Pairwise Adonis (R v4.2.3) were performed to assess statistical significance for beta-diversity. A significance threshold of P ≤ 0.05 or a Q ≤ 0.05 was used, with tendencies noted for values ≤ 0.10. DESeq2 was used to determine the differential abundance of ASVs between CON and AGO treatment groups across all time points (QIIME2 v2022.8; R v4.2.3) (Anders and Huber 2010).

Results

Sow lactation performance parameters

There was no significant impact of AGO supplementation on sow ADFI at any point of the study (Table 2). There were no differences among treatments in initial BW, BCS, or the Knauer Caliper measurement (Table 2). No differences were observed in BW, BCS, or the Knauer Caliper measures among treatments on d7 or at weaning (Table 2). While there were no differences in the change in BW, change in BW as a percentage of initial BW, or change in BCS, from d7 to weaning, there was a significantly lower loss of body condition as measured by the Knauer Caliper in AGO compared to CON (Table 2).

Table 2.

Effect of dietary essential oil supplement (CON vs AGO) on sow lactation performance and feed intake.

	Treatment Group1			P-value
	CON	AGO	SE	Treatment
No. of Sows	13	12	–	–
Parity	3.20	3.04	0.265	0.668
Day −14 Pre-farrowing
Sow BW, kg	241.9	250.4	8.31	0.459
Sow BCS2	3.3	3.5	0.20	0.476
Sow Caliper Measurement3	14.4	13.8	0.68	0.525
Day 7 Post-farrowing
Sow BW, kg	224.5	231.7	6.69	0.438
Sow BCS	3.1	3.1	0.29	0.841
Sow Caliper Measurement	14.5	13.3	0.71	0.257
Weaning
Sow BW, kg	221.5	226.6	5.60	0.525
Sow BCS	2.9	3.0	0.15	0.632
Sow Caliper Measurement	12.8	13.1	0.44	0.590
Change from Day 7 to Weaning4
Change in BW, kg	−5.1	−5.3	3.15	0.958
Percent Change in BW, %	−2.2	−2.2	1.29	0.983
Change in BCS	−0.1	−0.1	0.23	0.825
Change in Caliper  Measurement	−1.8	−0.1	0.43	0.010
Feed Intake
Week 1 ADFI, kg	4.13	3.99	0.375	0.672
Week 2 ADFI, kg	5.21	4.72	0.375	0.323
Week 3 ADFI4, kg	7.21	7.27	0.375	0.901
Overall ADFI4, kg	5.52	5.32	0.375	0.632

¹Treatment included sows that received either a basal diet that met or exceeded all NRC (2012) requirements (CON) or a basal diet + 200 ppm Agolin Pig (AGO).

²Body Condition Score (BCS) was assessed on a 1 to 5 scale with 1 being emaciated and 5 being overly fat.

³Sow Caliper measurement was measured using the Knauer Sow Caliper at the 12^th rib.

⁴Lactation length, number weaned, and parity class were used as covariates for all changes from Day 7 to Weaning, week 3 ADFI, and overall ADFI.

Piglet performance parameters

The average lactation length for AGO and CON was 19.1 ± 0.26 d (Table 3). The total number born alive, number of stillborns, and number of mummies did not differ between treatment groups (Table 3). The average number of pigs nursed post-crossfostering was 11.7 ± 0.41, and the average number of pigs weaned was 11.3 ± 0.30 (Table 3). There were no differences in initial piglet weights or weaning weights between AGO and CON (Table 3). Similarly, there were no significant differences in piglet ADG between AGO and CON (Table 3).

Table 3.

Effect of dietary essential oil supplement (CON vs AGO) on piglet performance.

	Treatment Group1			P-value
	CON	AGO	SE	Treatment
No. of Litters	13	12	–	–
Lactation Length, d.	19.2	18.8	0.28	0.345
No. of Total Born Alive	12.8	13.9	0.71	0.280
No. of Stillborns	0.7	0.8	0.28	0.664
No. of Mummies	0.8	0.5	0.29	0.362
No. Nursed2	11.8	11.5	0.41	0.488
No. Weaned	11.5	11.1	0.30	0.403
Average Piglet Weights3
Day 13, kg	1.5	1.5	0.25	0.914
Day 7, kg	2.8	2.6	0.43	0.239
Day 14, kg	4.5	4.2	0.66	0.243
Weaning, kg	6.0	5.6	0.84	0.303
Piglet Average Daily Gain5
Week 1, g/d	178	152	39.0	0.114
Week 2, g/d	249	232	40.5	0.334
Week 3, g/d	280	273	56.5	0.751
Overall, g/d	231	215	35.8	0.262
Average Litter Weights3
Day 13, kg	19.3	19.4	0.91	0.935
Day 7, kg	32.6	31.2	1.46	0.500
Day 14, kg	51.8	50.1	2.21	0.582
Weaning, kg	67.7	64.2	2.67	0.358
Litter Average Daily Gain5
Week 1, g/d	1894	1694	164.4	0.395
Week 2, g/d	2769	2669	148.0	0.630
Week 3, g/d	2952	2629	265.2	0.393
Overall, g/d	2510	2331	123.8	0.314

¹Treatment included sows that received either a basal diet that met or exceeded all NRC (2012) requirements (CON) or a basal diet + 200 ppm Agolin Pig (AGO).

²Number nursed was recorded at the completion of crossfostering by 48 h post-farrowing.

³Parity class was included in the model as a covariate for body weights on day 7, 14 and weaning. Lactation length was included in the model as a covariate for weaning weights.

⁴Day 1 weights were piglet and litter weights recorded within 24 h of farrowing at the time of processing as standard on farm practice at Purdue University.

⁵Parity class was included in the model as a covariate for all ADG. Lactation length was included in the model as a covariate for week 3 and overall ADG.

Milk composition

There were no significant differences between treatment groups in milk protein content throughout lactation (Table 4). Creamatocrit data showed no significant differences between treatments in fat content of colostrum or d3 milk samples, although the AGO group tended to have greater milk fat concentrations on d14 than the CON group (Table 4). There were no significant differences between treatments in IgA or IgG levels (Table 4).

Table 4.

Effect of dietary essential oil supplement (CON vs AGO) on milk composition.

	Treatment Group			P-value
	CON	AGO	SE	Treatment
No. of Sows	13	11	–	–
Day 01
Fat, %	10.3	10.4	0.86	0.893
Protein, %	8.7	9.9	1.21	0.483
IgG, mg/mL	43.0	51.6	8.88	0.518
IgA, mg/mL	8.7	6.2	1.79	0.189
Day 32
Fat, %	14.2	15.6	1.63	0.533
Protein, %	3.7	4.7	1.17	0.558
Day 143
Fat, %	11.0	13.1	0.75	0.062
Protein, %	3.1	3.8	1.21	0.692

¹Day 0 was within 2 h of the onset of farrowing and represented colostrum. Colostrum samples were collected from all teats and pooled to create a single homogenate sample.

²Day 3 was 3 days post-farrowing and represented transition milk. Milk samples were collected from 3 functioning mammary glands selected along the midline of the sow, and milk was extracted from each selected teat, measured in triplicates for each teat, and averaged.

³Day 14 was 14 days post-farrowing and represented lactation milk. Milk samples were collected from 3 functioning mammary glands selected along the midline of the sow, and milk was extracted from each selected teat, measured in triplicates for each teat, and averaged.

Microbiome

Species evenness for both CON and AGO was evaluated using the Pielou evenness index, which indicated relatively high and comparable evenness across all timepoints. There were no significant differences in species evenness between CON and AGO on Day −14 (Fig. 1A; P = 0.741), Day −7 (Fig. 1B; P = 0.789), Day 0 (Fig. 1C; P = 0.911), Day 8 (Fig. 1D; P = 0.469), or Day 16 (Fig. 1E; P = 0.230).

Figure 1.

Comparison of the Pielou Evenness of the Control (CON) and Agolin (AGO) treatments by day; 14 days prior to farrowing (A. Day −14; P = 0.741), 7 days prior to farrowing (B. Day −7; P = 0.789), Day 0, time of farrowing (C. Day 0; P = 0.911), 8 days post-farrowing (D. Day 8; P = 0.469), and 16 days post-farrowing (E. Day 16; P = 0.230).

Faith phylogenetic diversity index, an alpha diversity measurement of phylogenetic diversity, did not show significant differences between CON and AGO on Day −14 (Fig. 2A; P = 0.610), Day −7 (Fig. 2B; P = 0.518), Day 8 (Fig. 2C; P = 0.217), or Day 16 (Fig. 2D; P = 0.994). However, on Day 0, the AGO group tended to have greater phylogenetic diversity than the CON group (Fig. 2E; P = 0.099).

Figure 2.

Comparison of Faith Phylogenetic Diversity of the Control (CON) and Agolin (AGO) treatments by day; 14 days prior to farrowing (A. Day −14; P = 0.6100), 7 days prior to farrowing (B. Day −7; P = 0.5180), Day 0, time of farrowing (C. Day 0; P = 0.0995), 8 days post-farrowing (D. Day 8; P = 0.2170), and 16 days post-farrowing (E. Day 16; P = 0.9940). Dots represent outliers in the treatment groups.

Unweighted Unifrac is a measure of beta diversity that analyzes the presence/absence of bacteria considering phylogenetic diversity across communities. There were no significant differences in Unweighted Unifrac between CON and AGO on Day −14 (Fig. 3A; P = 0.906), Day −7 (Fig. 3B; P = 0.535), or Day 0 (Fig. 3C; P = 0.475). On Day 8, the AGO community composition was different than the CON group (Fig. 3D; P = 0.029) and tended to differ on Day 16 (Fig. 3E; P = 0.066).

Figure 3.

Comparison of Unweighted Unifrac analysis of the Control (CON) and Agolin (AGO) treatments by day; 14 days prior to farrowing (A. Day −14; P = 0.906), 7 days prior to farrowing (B. Day −7; P = 0.535), Day 0, time of farrowing (C. Day 0; P = 0.475), 8 days post-farrowing (D. Day 8; P = 0.029), and 16 days post-farrowing (E. Day 16; P = 0.066).

DESeq2 results indicated the differential abundance of bacteria between the AGO and CON groups at various time points (Fig. 4). A significant shift is noted if the log2fold change is greater than or equal to 2. Prior to treatment (d-14), Methanosphaera, Phascolarctobacterium, and two species of Treponema were significantly more abundant in AGO than CON (Fig. 4A). On Day -7, Methanobrevibacter, Treponema, and two unclassified Bacteroidales were more abundant in CON, while an unknown phylum, Prevotella, and an unclassified Clastridiales were more abundant in AGO (Fig. 4B). There were no significant differences in bacteria abundance across phylums between AGO and CON on Day 0. By Day 8 of lactation, several bacterial taxa showed differential abundance between AGO and CON groups most notably Clostridium, Gemmiger, Oscillospira, and Catebibacterium more abundant in the feces of CON sows while Fibrobacter and Phascolarctobacterium are more abundant in the AGO sows (Fig. 4C). Prevotella, Methanobrevibacter, and Streptococcus were more abundant in the CON group on Day 16 than in the AGO group while Treponema and Fibrobacter was more abundant in the AGO group (Fig. 4D).

Figure 4.

Results of DESeq2 analysis between AGO and CON treatments based on Phylum and Genus by day; 14 days prior to farrowing (A), 7 days prior to farrowing (B), 8 days post-farrowing (C), and 16 days post-farrowing (D). There were no differences between AGO and CON on Day 0, time of farrowing. Differences are reported as a log2FoldChange. A positive number indicates an increase in population in the CON sows, while a negative number indicates an increase in population in the AGO sows.

Discussion

The modern sow has been selected for high prolificacy, producing greater numbers of piglets with higher lean tissue deposition (Ward et al. 2020). Modern sows are often unable to meet their nutritional requirements with voluntary feed intake leading to the mobilization of body reserves (Silva et al. 2018) and increase in oxidative stress during lactation (Berchieri-Ronchi et al. 2011). Additionally, the increasing pressure to reduce antimicrobial use in the swine industry has led to the evaluation of a variety of natural alternatives such as EOs to control microbial populations and improve feed intake in the sow. This study was designed to evaluate the effects of Agolin Pig, a blend of microencapsulated essential oils (clove, eugenol, thymol, piperine, and geraniol), on sow and litter performance and fecal microbiota composition during lactation.

Previous studies have shown conflicting results on the impacts of EOs on sow lactation performance and piglet performance. A study using phytogenic feed additives with or without additional dietary fiber in gestating and lactating sows found that supplementation of 1 ppm Scutellaria baicalensis and Lonicera japonica mixed extracts from late gestation to weaning enhanced the number of piglets born alive, litter birth weight, litter weight gain, and average daily feed intake of sows during lactation while decreasing the incidence of diarrhea of piglets (Wang et al. 2022). Other studies have shown an increase in feed palatability and digestibility in pigs fed diets containing EO supplements (Janz et al. 2007; Ivanova et al. 2021; Papatsiros et al. 2024). Conflicting studies have shown no effect of EO supplementation on performance and feed intake in various stages of production (Neill et al. 2006; Stelter et al. 2013; Z. Zeng et al. 2015). Therefore, it is not entirely unexpected that the supplementation of EOs in this preliminary study did not affect sow lactation feed intake and piglet performance. However, a published abstract at the Pig Research Summit in Denmark is one of the limited research studies supplementing Agolin Pig at 100 ppm to sows from d107 of gestation through lactation resulted in 5.2% higher litter ADG compared to control sows with no effect on individual piglet growth performance (Spengler et al. 2023). The study conducted by Spengler et al. (2023) differed from the current study by utilizing a lower dose of Agolin Pig for a shorter period of time that likely contributed to the difference in results. Future studies should explore dietary inclusion levels and timings of Agolin Pig to identify the optimal protocol for this EO blend.

The Knauer Caliper has proven to be more accurate than subjective BCS by visual assessment in detecting changes in body condition with calipers identifying reductions in sow body condition that could be missed by BCS or weight measurements (Li et al. 2021). Ideal Knauer Caliper measurements for maintenance of body condition and reproductive performance should fall between 12.5 and 14.0 units in pre-farrowing sows across parities (Li et al. 2021). In this study, all sows were within the ideal range or slightly above at 14 days prior to farrowing and remained in the ideal range at weaning. While there were no significant differences in BW, BCS, or Knauer Caliper measurements among treatment groups, there was a significant lower change of Knauer Caliper measurement from d7 to weaning in the AGO group compared to the CON group, indicating that AGO potentially mitigated the loss of body condition as measured by the Knauer Caliper. These findings are similar to previous research, which resulted in a numerical decrease in body condition loss in those supplemented with 100 ppm of Agolin Pig (6.0% loss in BCS) compared to control sows (8.2% loss in BCS; Spengler et al. 2023). A greater sample size in future studies may be able to detect a greater impact of AGO supplementation on sow body condition maintenance during lactation.

The fat content in the colostrum (d0), transition milk (d3), and mature milk (d14) were higher in the present study compared to previously reported values by Theil et al. (2014). The study reported that late colostrum contains ∼7% fat while we found colostrum to have about 10% fat. Transition milk at 72 h was reported to contain about 10% fat while we found it to contain 14–15% fat. Mature milk at 17 d had ∼8% fat while we found 11–13% fat. Additionally, we observed similar protein concentrations in colostrum as those reported by Theil et al. (2014), but transition and mature milk yielded lower protein concentrations than those previously reported (3.7–4.7% vs. 6.1% and 3.1–3.8% vs 4.7%, respectively). These differences in milk composition analyses may be related to different analysis methodologies and timings employed. Previous studies have shown that oregano EO supplementation increased IgG, IgA, and T-lymphocyte concentrations in the milk of lactating sows (Ariza-Nieto et al. 2011; Omonijo et al. 2018). However, in this study, there appeared to be no significant effects of AGO supplementation on colostrum.

It has been frequently documented that weaned piglets treated with various essential oils exhibit an increase in Lactobacillus species and a decrease in the presence of Escherichia coli in their feces (Castillo et al. 2006; Wei et al. 2017; Z. Zeng et al. 2015). However, few studies have explored the impact of EOs on the sow fecal microbiota. In a study where sows were supplemented with an EO blend at 700 ppm during gestation or both gestation and lactation, no significant differences in alpha-diversity indices between the control group and the group fed EOs was measured prior to move-in to the farrowing house (Nowland et al. 2021). However, the study resulted in differences in microbial compositions with higher levels of unclassified p253418B5, unclassified Bacteria, Enterococcus, Sporobacter, Succinispira and the archaea Methanobrevibacter in control sows and Roseburia, Subdoligranulum, Lactonifactor, Oscillospira, Coprococcus, Pediococcus, p75a5, CF231, Prevotella, Ruminococcus, unclassified S247, and Butyrivibrio higher in EO-treated sows during gestation and lactation.

Throughout the study, several specific shifts in microbial taxa between the treatment groups provided insight into the potential efficacy of Agolin Pig. Although sows were randomly assigned to treatment groups based on parity and genetics, there were initial differences in their fecal microbiota. The AGO group had higher levels of Methanosphaera, Treponema, and Phascolarctobacterium compared to the CON group, with log2 fold differences greater than 2.5, likely due to random variation. Over the course of the study, Treponema levels fluctuated inconsistently in both groups. Treponema, which is commonly found in mature pigs, plays a role in cellulose and lignin degradation (Niu et al. 2015) and has also been associated with normal heat cycles in sows (Wang et al. 2021).

By 7 days prior to farrowing, Methanobrevibacter showed over a 3 log2-fold increase in abundance in the CON group. Methanobrevibacter is a dominant archaea genus found in the healthy pig hindgut (Gresse et al. 2019). While Prevotella abundance increased in the AGO group by Day -7, this shift was inconsistent across the study and was seen in both groups on Day 8 and in the CON group only on Day 16 of lactation. Prevotella has been positively correlated with improved growth performance, immune function, and reproductive cyclicity in swine (Amat et al. 2020).

By Day 0, no significant microbial differences were detected between the AGO and CON groups. However, by Day 8, several taxa showed notable changes, likely driven by the stress of farrowing and the shift from restricted to ad libitum feeding, which can alter gut microbiota (Huang et al. 2019). Notable shifts included an increase in Clostridium, Gemmiger, Oscillospira, and Oribacterium in the CON group. Clostridium contains several virulent strains that are linked to diarrhea and necrotic enteritis (Baker et al. 2010), while Oscillospira has been associated with negative intestinal interactions and reduced cyclicity in sows (Wang et al. 2021). Gemmiger, which produces formic and butyric acids, is more commonly found in chickens (Gossling and Moore 1975), and Oribacterium is linked to increased piglet body weights and higher-producing sows (Wang et al. 2019). In contrast, the AGO group showed an increase in Phascolarctobacterium by Day 8, which has been associated with improved reproductive cyclicity in sows (Gaukroger et al. 2021). While these genera can provide us with potential high-level mechanisms impacted by AGO, future research is needed to understand the species-specific impact within the microbiota that was not evaluated in this pilot study.

These microbial shifts suggest that the AGO group experienced a more beneficial change in fecal microbiota compared to the CON group, as the latter showed an increase in potentially harmful bacteria by Day 16. Notably, the CON group exhibited a significant increase in Streptococcus, a pathogenic bacterium (Fittipaldi et al. 2012), while the positive changes seen in Prevotella and Methanobrevibacter were inconsistent across groups. Additionally, the AGO group showed a similar increase in Fibrobacter on Days 8 and 16, a bacterium essential for fiber digestion in swine. Previous research has also observed an increase in Fibrobacter in sows supplemented with oregano essential oil (Hall et al. 2021). As the understanding of dietary fiber and sow microbiota continues to expand, research has indicated that increasing dietary fiber leads to an increase in fiber-digesting bacteria and subsequent enhancement of reproductive health through improved intestinal health, enhanced immune system function, regulation of lactation and production performance, and modification of short-chain fatty acid production (Tian et al. 2020). Future research is needed to understand the potential benefits of Agolin Pig in improving sow fecal microbiota and reproductive health in sows.

Although blood samples were not collected in this study to measure estrogen levels in the sows, future analysis using Agolin Pig could investigate potential correlations between fecal microbiome composition and reproductive hormone cyclicity, particularly evaluating differences between sows and gilts. This research is especially relevant as a previous study supplementing sows with 100 ppm of Agolin Pig in late gestation and throughout lactation led to improvements in pregnancy rates post-weaning (88.2%) compared to control sows (81.3%; Spengler et al. 2023). Future research is needed to evaluate the long-term impact on reproduction specifically through sow reproductive cyclicity and subsequent pregnancy performance.

In general, the microbial communities in AGO and CON groups had mostly inconsistent shifts across the study, but it appears that CON sows had increased abundance of bacteria genera that may contain pathogenic species compared to AGO by Days 8 and 16 post-farrowing. This suggests that Agolin Pig had a small, positive impact on sow fecal microbiota communities during late gestation and lactation at the concentration fed. However, the observed shifts in Treponema and Prevotella at different time points warrant further research to determine if these were pathogenic species and understand the potential implications for sow health.

Conclusion

Overall, in this preliminary study, Agolin-Pig significantly reduced sow loss of body condition as measured by the Knauer Caliper with no other differences observed in any other measures of body condition and body weight. While not statistically significant, there was a numerical trend for an increase in fat, Kcal, and protein concentrations in the milk in the AGO group compared to the CON. There were no differences in ADFI of sows among treatment groups and piglet growth performance did not differ among groups. Sows supplemented with AGO potentially had added protection from the colonization of Clostridium, Oscillospira, and Streptococcus, as these genera contain potentially harmful bacterial species and were more abundant in the CON as measured in DESeq2. However, there was only a significant difference observed in phylogenetic diversity between treatments on Day 8 post-farrowing and a tendency on Day 16 with all other measures of alpha and beta diversity resulting in no differences between treatment groups. In conclusion, this preliminary study indicates the potential of Agolin Pig to potentially mitigate the loss of body condition and alter the diversity of sow fecal microbiota communities during lactation.

Glossary

List of Abbreviations

ADFI

average daily feed intake

ADG

average daily gain

AGO

Agolin

BCS

Body Condition Score

BSA

bovine serum albumin

BW

body weight

Ca

calcium

CON

control

CF

crude fat

CP

crude protein

d

day

EO

essential oil

EU

European Union

g

gram

IgA

Immunoglobulin A

IgG

Immunoglobulin G

kg

kilogram

mg

milligram

No

number

P

phosphorous

Funding

This study was funded by Smithfield Foods with the product provided by FEEDWORKS USA, Ltd.

Author contributions

Individual author contributions to this research article are as follows: Conceptualization and design: A.E.D., B.T.R., K.R.S., M.A.P., and M.J.H; methodology: A.E.D., A.E.F., B.T.R., K.H., K.R.S., M.A.P., M.J.H., T.A.J., T.J.S., and W.X.D.; statistical analysis: A.E.F., B.T.R., and K.R.S.; investigation: A.E.F., B.T.R., K.H., K.R.S., T.A.J., T.J.S., and W.X.D.; writing-original draft preparation: A.E.F., B.T.R., and K.R.S.; writing-review and editing: A.E.D., A.E.F., B.T.R., K.H., K.R.S., M.A.P., M.J.H., T.A.J., and T.J.S. All authors have read and agreed to the published version of the manuscript.

Conflicts of interest

Alexandra Fisk is currently employed by Ralco Agriculture, a competitor of Agolin. However, the research presented in this manuscript on Agolin Pig was conducted prior to this employment. Ralco Agriculture had no role in the study design, data collection, analysis, interpretation, or writing of this paper.