Evaluation of oxidized beta-carotene on sow and piglet immune systems, sow reproductive performance, and piglet growth

Abstract

This study aimed to determine if supplementation of oxidized-beta carotene (OxC-Beta) improved sow reproductive performance, litter growth performance, vitamin A status, and ability to alter immune cells abundance in sows and piglets, subsequent litter performance, and nursery growth performance. On approximately day 60 of gestation and through the lactation period, 194 sows (blocked by parity) were assigned to a common gestation diet or the common diet supplemented with 80 ppm oxidized beta-carotene (OxC-Beta, Aviagen, Ottawa, ON, Canada). A subset of sows (N = 54 per treatment) were sampled for blood and body weight recorded at the beginning of the study, farrowing, and weaning. A blood sample was taken from a subset of piglets at birth and weaning, and all piglet weights were recorded. Blood was analyzed for vitamin A as retinol concentrations and immunoglobulin G (IgG) and immunoglobulin M (IgG) levels were assessed from the sow’s blood. Twelve pigs (N = 6 per treatment) were euthanized at birth and weaning. The livers were collected and analyzed for the Kupffer cell phagocytic activity through flow cytometry. Whole blood was analyzed via flow cytometry for cluster of differentiation (CD335, CD8, and CD4). Colostrum during farrowing and milk at weaning were analyzed for IgG and IgA concentrations. Data were analyzed via SAS 9.4 using MIXED and frequency procedures where appropriate. No differences (P > 0.05) between dietary treatments were observed in sow reproductive performance, feed intake, wean to estrus interval, or piglet growth performance. No differences (P > 0.05) were observed in the plasma or liver for vitamin A. No differences (P > 0.05) were observed in the composition of the colostrum or milk. No immunological differences (P > 0.05) were observed in the piglets’ liver and blood or sow antibodies in colostrum and milk. The supplementation of OxC-Beta did (P < 0.05) decrease IgM and tended (P < 0.10) to decrease IgG in sow plasma. No differences (P > 0.05) were observed in the reproductive performance of subsequent litter information from the sows. Gilt litter weaning weight and feed intake were reduced (P < 0.05) compared to sow performance. In conclusion, the supplementation of OxC-Beta at 80 ppm from day 60 of gestation through lactation does not affect the reproductive performance of sows, litter growth performance, vitamin A status, piglet immune status, and antibodies or composition in colostrum and milk.

Supplemented oxidized beta-carotene did not affect the reproductive performance of sows, litter growth, or measured immune function of piglets.

Introduction

The reproductive performance of sows is one of the top economic factors in swine production (Coffey and Britt, 1993); thus it is essential to give each sow the best chance of having a productive and healthy litter. Progeny of primiparous sows are associated with having a lighter birth weight (Hendrix et al., 1978; Miller et al., 2013; Craig et al., 2017) and lighter weaning weight (Wilson and Johnson, 1980; Carney-Hinkle et al., 2013; Craig et al., 2017) than multiparous sows progeny. A decrease in growth rates from the progeny of primiparous sows could be attributed to an overall reduced health status resulting from primiparous sows having reduced passive immunity and increased susceptibility to pathogens (Carney-Hinkle et al., 2013). Primiparous sows do not have as robust of an immune system as multiparous sows, meaning they are more vulnerable to infections that may already be present or those that may pass through a breeding herd (Scruton et al., 2002).

Colostrum is the first milk that is secreted from the sow for 0 to 24 h postpartum and contains high levels of immunoglobins and nutrients, such as proteins, fats, and lactose, that are essential for piglet survival. The major proteins that are found in colostrum are identified as immunoglobins, mainly immunoglobin G, which are needed by the piglets as they are born immunologically naïve as antibodies do not transfer through the placenta (Alexopoulos et al., 2018). Immunoglobulins in the colostrum are essential as they provide the first source of immune protection to the piglet and can be limited due to colostrum intake, absorption, and the immunoglobulins that the sow provides (Varley, 1995). Furthermore, primiparous sows are reported to have lower levels of immunoglobulins in their colostrum and milk compared to multiparous sows (Farmer and Quesnel, 2009). Absorption of these immunoglobulins is most critical prior to gut closure when immunoglobulins can easily be uptaken into the lymphatic system of piglets (Varley, 1995). Gut closure begins within 24 h of birth and is complete by 48 h after birth to prevent pathogen uptake (Varley, 1995). Milk contains the same nutrient components of colostrum, but at different concentrations. For example, in milk, the most densely secreted immunoglobulin is immunoglobulin A, instead of G. Nutritional strategies have been studied to improve items such as milk composition, oxidative stress, body condition score, and immunity passed from sow to piglet. One potential strategy is the supplementation of a carotenoid to help improve the immune function of the sow.

One of the most known functions of carotenoids, particularly beta-carotene, is being a precursor for vitamin A (Combs, 2012). Vitamin A is essential for immune cell modulation (Huang et al., 2018), epithelial cell differentiation (Barbalho and Pescinini-Salzedas, 2019), and embryonic development (Combs, 2012). Thus, vitamin A plays a key role not only during pregnancy supporting embryonic development, but vitamin A is also essential for the immune system. However, some benefits of carotenoids independent of being a vitamin A precursor also exist. In humans, carotenoids have been documented to provide health benefits such as preventing cataracts, macular degeneration, and improving immunomodulation (Mayne, 1996). Additionally, beta-carotene is involved in the immune system and modulating oxidative stress (Johnston et al., 2014). Additionally, beta-carotene and vitamin A injections given to sows result in an improvement in reproductive performance (Brief and Chew, 1985; Coffey and Britt, 1993). Recent research has shown that beta-carotene will undergo spontaneous and complete oxidation, yielding a copolymer product with immunomodulating functions (Burton et al., 2016). One proposed strategy is that oxidized beta-carotene can improve the immune status of sows, in turn helping improve piglet health (Jun et al., 2021), and ostensibly improved growth performance. It was hypothesized that supplemented oxidized beta-carotene during gestation and lactation can prime the sow’s immune system (Johnston et al., 2014), which consequently would benefit progeny growth and immune status. Thus, this study aimed to feed gestating sows an oxidized beta-carotene (OxC-beta) product through lactation to determine if using an oxidized beta-carotene product can improve piglet birth weight, immune status, and reduce preweaning mortality.

Materials and Methods

All procedures in this experiment adhered to guidelines for the ethical and humane use of animals for research and were approved by the Institutional Animal Care and Use Committee at Iowa State University (IACUC #20-059).

Animals

A total of 194 sows (Topigs Norsvin TN20, TN70, and Z-line, Vught, Noord-Brabant, the Netherlands) in two groups were assigned to one of two dietary treatments: a common diet (CON; N = 99) or the test diet (OxBC; N = 95) and blocked by parity (gilts (P0) N = 47; first parity (P1) N = 38; and second parity and older (P2+) N = 109). During gestation, sows were group housed by treatment in a total of 24 5.49 × 2.44 m² pens that had slatted floors with one wall-mounted water nipple per pen with 8 to 10 sows per pen. All pens were floor fed. Sows were moved into the farrowing house approximately 3 d before the predicted farrow date. Farrowing stalls were 2.13 × 1.52 m², and the sow’s fixed stall was 0.56 m wide. The commercial farm was considered stable for porcine reproductive and respiratory syndrome virus, having measurable antibodies from a wild-type exposure 6 mo prior to the start of the trial [herd category positive stable (II-A), Holtkamp et al., 2011]

Diets

Animals were fed one of two diets starting on day 60 of gestation. A control diet of a corn and soybean meal base (Table 1), or a test diet, which was the control diet supplemented with 80 ppm OxC-Beta (10% oxidized beta-carotene, Avivagen, Ottawa, ON, Canada). All diets were manufactured on the farm and were formulated to meet or exceed NRC (2012) requirements for both gestating and lactating swine. Sows began receiving dietary treatments at approximately day 60 of gestation and received the dietary treatment through the lactation period. In gestation, sows were fed according to the average body condition score of the pen in order to obtain an ideal body score of 3 (Coffey et al., 1999). After the sow had farrowed, the feed was gradually increased 0.9 kg/d from an initial 2.05 kg/d until the sow could no longer consume all the feed in her feeder.

Table 1.

Formulation, calculated, and analyzed gestation and lactation control and test diets (control supplemented with oxidized beta-carotene; OxBC)

Diet	Gestation control	Gestation OxBC	Lactation control	Lactation OxBC
Ingredient, %
Corn	82.25	82.25	70.45	70.45
Soybean meal (46.5% crude protein)	14.25	14.25	25.45	25.45
Vitamin and trace mineral premix1	3.50	3.50	3.50	3.50
L-Lysine HCl	–	–	0.34	0.34
L-Threonine	–	––	0.16	0.16
DL-Methionine	–	–	0.10	0.10
L-Tryptophan	–	–	0.02	0.02
Oxidized beta-carotene	–	0.008	–	0.008
Calculated composition, %
Dry Matter	87.8	87.8	88.8	88.8
Crude protein	12.5	12.5	17.3	17.3
Calcium	0.75	0.75	0.81	0.81
Available phosphorus	0.45	0.45	0.46	0.46
SID2 Lysine	0.55	0.55	1.05	1.05
Analyzed composition, %
Dry Matter	86.41	87.24	86.88	86.88
Ash	3.39	4.31	4.84	4.93
Energy, kcal/kg	3,731	3,688	3,735	3,716
Nitrogen	2.00	1.98	2.76	2.77
Crude protein	12.49	12.36	17.27	17.33

¹Vitamin and trace mineral premix; limestone 32.85%, salt 14.30%, L-Lysine 1.40%, monocalcium phosphate (21%) 34.30%, corn distillers 1.85%, choice white grease 1.00%, and micromineral mix = calcium (minimum 6.03%, maximum 7.23%), zinc (minimum 33,000 ppm), selenium (minimum 59.99 ppm), and phytase (minimum 45,360 FTU/LB).

²SID, Standardized ileal digestible.

Feed chemical analysis

A feed sample (approximately 200 g) was taken from each mixed batch by taking a sample dispensed from the bin at the start of the new batch and stored at −20 °C for later analysis. Feed samples from all batches of each diet were pooled (roughly 200 g total), ground to 1 mm particle size, using a Wiley Mill (Variable Speed Digital ED-5 Wiley Mill; Thomas Scientific, Swedesboro, NJ). Ground samples were analyzed in duplicate for gross energy, dry matter (method 930.15; AOAC, 2007), ash (method 942.05; AOAC, 2007), and nitrogen (method 990.03; AOAC, 2007; TruMac; LECO Corp., St. Joseph, MI). For calibration, an ethylenediaminetetraacetic acid sample (LECO Corp.; 9.56% nitrogen) was used as the standard. Crude protein was calculated as nitrogen × 6.25. Gross energy was determined using an isoperibolic bomb calorimeter (model 6200; Parr Instrument Co., Moline, IL). Benzoic acid (Parr Instrument Co.; 6,318 kcal GE/kg) was used as the standard for calibration.

Sample collection

A subset of sows (N = 112) was randomly selected to be sampled throughout the study (P0 = 12; P1 = 9; P2+ = 35 per diet) with the number of sows in each parity to be representative of the parity distribution on the farm. The subset of sows was randomly selected to be representative of the sows across parities and across treatments. Sow weight was estimated and recorded via heart girth tape (Groesbeck et al., 2002) from the subset of sample sows at the start of the study, when the sows were moving into the farrowing house, and at weaning. All litter weights were recorded within 24 h of farrowing. Cross fostering within a treatment occurred within 24 h of the sow farrowing.

Flow cytometry

Twelve presuckle piglets (one per litter) per diet were randomly selected solely based on presuckle status for tissue collection and immune system assessment at birth following humane euthanasia via blunt force trauma. The previous was repeated at weaning, using pigs (N = 12 per diet, a total of 48 piglets over the course the study) from the same litter that was randomly selected at birth. Immediately postmortem, a blood sample was collected from the hepatic vein into a dipotassium ethylene diamine tetraacetic acid blood tube (BD Vacutainer, Franklin Lankes, NJ). The right lobe of the liver was collected for analysis of phagocytic activity measured via flow cytometry. Livers were transferred in Gibco Hanks’ Balanced Salt Solution (1X) with no calcium chloride, no magnesium chloride, and no magnesium sulfate (Fisher Scientific, Waltham, MA, US) on ice. The left lobes of the livers were collected, transferred on ice, and stored at −80 °C until further analysis. The left lobes of the livers were submitted to the Iowa State University Veterinarian Diagnostics Laboratory (Ames, IA) for vitamin A analysis as described in Greiner et al. (2022). Vitamin A in the liver was measured by converting retinyl palmitate to retinol, and then measuring all retinol levels in the sample. All vitamin A metabolites measured will be referred to as “vitamin A” in the results and discussion.

Blood was kept chilled and occasionally rocked until analysis, approximately 1 h later, at which point 50 µL of whole blood was diluted with 50 μL phosphate buffer saline (PBS; room temperature, pH 7.4; Fisher Scientific, Waltham, MA). Mouse antipig cluster of differentiation (CD) 335 with a allophycocyanin fluorescent label, mouse antipig CD8 alpha with a phycoerythrin fluorescent label, and mouse antipig CD4 with a fluorescein isothiocyanate fluorescent label (CAT: MCA5972APC, MCA6048PE, MCA6045F, respectively; Bio-Rad Laboratories, Inc., Hercules, CA, US) were added, respectively, at 10 μL per antibody, covered in foil, and allowed to incubate for 15 min at room temperature (roughly 20 °C). Cells were then washed with 1 mL of PBS. The samples were then lysed with BD FACS Lysing Solution 10X (Cat: 349202, BD Bioscience, Franklin Lakes, New Jersey, US) diluted to 1X with room temperature MilliQ water. Samples were incubated with lysing solution for 5 min at room temperature (roughly 20 °C) while covered in foil. The lysing solution was spun off the cells. The cells were then resuspended in 500 µL of BD Stabilizing Fixative 3X (Cat: 338036, BD Bioscience, Franklin Lakes, New Jersey, US) diluted to 1X with room temperature (roughly 20 °C) MilliQ water. Samples were incubated with the fixative for 1 h at 4 °C while covered in foil. The fixative solution was spun off, and cells were resuspended in 175 µL PBS, transferred to Falcon Round-Bottom Polystrene 5 mL Test Tubes (Corning Life Sciences, Corning, New York, US), covered in foil, and stored at 4 °C. Samples were analyzed via flow cytometry within 48 h of being fixed.

Flow cytometry phagocytic activity in the liver

Approximately 3 g of the right lobe of the liver was mechanically digested to obtain a single-cell suspension. Once obtained, cells were suspended in approximately 12 mL of Gibco advanced Roswell Park Memorial Institute medium 1640 (1X), reduced serum medium with nonessential amino acids, 110 mg/L sodium pyruvate, and no L-glutamine (Fisher Scientific, Waltham, MA, US). The solution was subsequently transferred to a petri dish for incubation in a humified 39 °C, 5% CO₂ incubator for 1 h and 40 min to isolate Kupffer cells via adhesion. Media was then removed from the petri dish, and cells were lifted from the petri dish and resuspended in roughly 3 mL of PBS. Isolated Kupffer cells were subjected to analysis of phagocytic activity via a Phagocytic Assay Kit (IgG with a phycoerythrin fluorescent label, Cat: 600540, Cayman Chemical, Ann Arbor, MI, US), where 1 µL of fluorescent-labeled beads were added to 499 µL of Kupffer cells suspension and allowed to incubate in a humified 39 °C, 5% CO₂ incubator for an hour. Cells were spun down to remove supernatant and then were resuspended in 500 µL of BD Stabilizing Fixative 3X (Cat: 338036, BD Bioscience, Franklin Lakes, New Jersey, US) diluted to 1X with room temperature MilliQ water. Samples were incubated with the fixative for 1 h at 4 °C while covered in foil. The fixative solution was spun off, and cells were resuspended in 175 µL PBS, transferred to Falcon Round-Bottom Polystrene 5 mL Test Tubes (Corning Life Sciences, Corning, New York, US), covered in foil, and stored at 4 °C. Samples were analyzed via flow cytometry within 48 h of being fixed.

Immunoglobulin analysis via enzyme-linked immunosorbent assay

Immunoglobulins (Ig) G and M were analyzed in sow plasma via enzyme-link immunoassay. Polystyrene Plates (CAT: 15041; ThermoFisher Scientific, Waltham, MA, US) were washed four times with wash solution (PBS with 0.5% tween 20, room temperature, pH 7.4), then coated with 100 μL capture antibody (IgG CAT: AAI41; IgM CAT: AAI48; Bio-Rad Laboratories, Inc., Hercules, CA, US) at a concentration of 12 µg/mL IgG or 7.5 µg/mL IgM and 100 μL coating buffer (CAT: CB01100; ThermoFisher Scientific, Waltham, MA, US) for 1 h at room temperature. Plates were then washed three times with a wash buffer and plated with 100 μL of blocking agent (CAT: DS98200, ThermoFisher Scientific, Waltham, MA, US) and left overnight at room temperature. Plates were washed three times with a wash buffer. Samples and a standard curve were then plated (100 μL) and left to incubate for 1 h at room temperature (approximately 20 °C). Every plate had a standard curve to account for differences across plates. The standard curve was developed using a pig IgG (CAT: I4381-10MG; Millipore Sigma; Burlington, MA) or pig IgM (CAT: 5276-6504; Bio-Rad Laboratories, Inc., Hercules, CA, US), respective to the assay being conducted. The native pig antibodies were diluted to 1000 ng/mL, and then serial diluted to 333.33, 111.11, 37, 12, 4, 1 ng/mL in order to develop the standard curve. A sample of the diluent (PBS) was also included as a blank. All samples were diluted 1:500,000 to ensure the measured absorbance could be captured by the standard curve. After samples were incubated, the plate was washed three times with wash buffer. A secondary antibody (IgG CAT: AA141P; IgM CAT: AAI48P Bio-Rad Laboratories, Inc., Hercules, CA, US) was plated at 100 μL per well at a concentration of 0.2 ug/mL for both IgG and IgM and left covered at room temperature for 1 h. Plates were washed three times before being coated with 100 μL 3,3ʹ,5,5;-tetramethylbenzidine chromogen solution (CAT: 002023, ThermoFisher Scientific, Waltham, MA, US) and incubated covered at room temperature for 20 min. A solution of 0.16 M sulfuric acid was added at 100 μL to stop the reaction. Plates were then read (Cytation 5 Hybrid Multi-Mode Reader, Biotek Instruments Inc., Winooski, VT) at a wavelength of 450 nm, and a coefficient of variation of under 8% between duplicates was deemed acceptable.

Colostrum and milk analysis

Approximately 35 mL of colostrum was collected during farrowing, and milk samples were collected at weaning. Piglets were removed from the sow for 1 h before 1 cc of oxytocin (MWI Veterinary Supply Inc., Boise, ID) was administered intramuscularly near the vulva. The underlines of sows were cleaned with a chlorohexidine scrub (Patterson Veterinary Supply, Greely, CO) and a clean paper towel before milk extraction. A clean gloved hand was used to expel milk from the second through fourth anterior teats (pooled) on both sides of the underline.

Colostrum and milk samples were sent to Iowa State University Veterinarian Diagnostic Laboratory (Ames, IA) for antibody analysis via enzyme-link immunoassay. Colostrum was diluted at 1:1,000,000 for IgG analysis and 1:20,000 for IgA analysis, while milk was diluted at 1:5,000 for IgG analysis and 1:2,000 for IgA analysis. Dilutions were performed so that sample concentrations fit within the ranges of the standard curve. Colostrum and milk composition were analyzed by MQT Lab Services (Kansas City, MO) for fat, lactose, total solids, true protein, and urea.

Record collection

Piglet sampling age was normalized based on when the last sow farrowed in that all piglets within a group had weaning samples collected at the same age (piglet age at weaning = 20 and 22 d for each group, respectively). At weaning, all litter weights and the weight of the sampled sows were recorded. Sows were monitored after weaning for estrus and wean-to-estrus interval (WEI) was recorded. Pregnancy status after post-weaning insemination and subsequent farrowing performance were also recorded. Piglets from the second group of sows were followed into the nursery for general performance records. All piglets received a common diet in the nursery. Piglets were housed with other piglets from the same dietary treatment within a nursery room. At the end of the nursery period, approximately 6 wk, a group weight was obtained via a truck scale as the pigs were transported from the farm.

Plasma collection

A plasma sample was taken on day 0 and at weaning via sterile, jugular venipuncture and again at farrowing via ear vein from the subset of sample sows, respectively. A plasma sample from one piglet per sample sow was collected at birth and weaning. At birth, blood was collected from the umbilical cord of presuckle piglets, and at weaning blood was collected via jugular venipuncture. A plasma sample was taken from the umbilical cord of a presuckle piglet and via jugular venipuncture at weaning from each subset sow litter, respectively. After all blood was collected, it was centrifuged (2,000 × g for 10 min at 4 °C; Sorvall Legend XFR, ThermoFisher Scientific, Waltham, MA, US) to separate plasma from red blood cells. The plasma was analyzed for vitamin A concentration as the metabolite retinol at the Iowa State University Veterinary Diagnostics Laboratory (Ames, IA) via ultrahigh-pressure liquid chromatography using methods described in Greiner et al. (2022).

Statistical analysis

All data were analyzed using the MIXED procedure in SAS 9.4 (Statistical Analysis System, Cary, NC) for all performance measures, vitamin A status, flow cytometry, colostrum and milk antibodies, colostrum and milk composition, and sow plasma immunoglobulins. The FREQ procedure in SAS 9.4 was utilized for the pregnancy status of the sow after rebreeding. The central limit theorem was utilized for normality due to the large sample size per dietary treatment. Subsequent nursery data were not statistically analyzed due to a lack of power (N =1). In the mixed model, sow was the experimental unit. Fixed effects were dietary treatment, parity, and farrowing room with a random effect of sow. All parameters of interest were analyzed for effects of dietary treatment, parity, dietary treatment by parity interaction, and dietary treatment by farrowing room interaction. Means were reported from LSMeans using a Tukey adjustment. P-values ≤ 0.05 were considered significant, while 0.05 < P-values ≤ 0.10 were considered a tendency.

Sows were removed from the study for reasons such as coming back into heat, abortion, death, and sows not producing milk for piglets during lactation. Litters were removed from performance analysis if litter mixing occurred in the farrowing house and unaccounted for piglet movement between litters.

Results

Analyzed dry matter and crude protein of the diets were similar to the calculated values (Table 1) for both gestation and lactation diets. Flow cytometry parameters of the piglets’ immune system were not different (P ≥ 0.61) between dietary treatments (Table 2). However, phagocytic activity tended (P = 0.05) to be higher at birth (2.85% of analyzed cells) than at weaning (0.33% of analyzed cells), and CD4, CD8 alpha, and CD335 receptors on cells differed significantly (P < 0.01) across time points. Receptors for CD4 were higher at birth (14.25% of analyzed cells) than at weaning (8.62% of analyzed cells). Receptors for CD8 alpha were higher at weaning (44.38% of analyzed cells) than at birth (5.83% of analyzed cells). Receptors for CD335 were higher at weaning (10.02% of analyzed cells) than at birth (3.92% of analyzed cells).

Table 2.

Effects of dietary treatment of maternal nutrition supplemented with oxidized beta-carotene (OxBC) on piglet Kupffer cell phagocytic activity and CD4, CD8 alpha, and CD335 cell populations analyzed via flow cytometry at birth and weaning

Time	Birth		Weaning			P-value1
Diet	Control	OxBC	Control	OxBC	SEM	TRT	Time	TRT xTime
Parameter2
Phagocytic activity in Kupffer cells	3.10	2.61	0.36	0.29	1.310	0.8432	0.0524	0.8660
CD4	14.30	14.20	9.45	7.79	1.993	0.6828	0.0061	0.6785
CD8 alpha	6.97	4.70	44.24	44.52	5.134	0.7920	<0.0001	0.7220
CD335	3.34	4.49	9.62	10.44	1.396	0.5360	<0.0001	0.8928

¹TRT, dietary treatment

²Kupffer cells were isolated from piglet liver. Receptors for CD4, CD8 alpha, and CD335 were from whole blood.

The IgG concentration in sow plasma tended (P < 0.10) to be decreased for sows fed OxC-Beta (89.74 vs. 125.52 ng/mL, respectively, Table 3). There was also a significant effect of time, in which IgG was higher during gestation than at ­farrowing. Sows fed OxC-Beta had a decreased (P < 0.01) concentration of IgM throughout the study (20.27 vs. 29.58 ng/mL, respectively, Table 3). IgM concentration was significantly different (P < 0.05) across time, with concentrations highest at gestation and lowest at weaning.

Table 3.

Dietary effects of oxidized beta-carotene (OxBC) on immunoglobulins¹ (IgG and IgM) in sow plasma at gestation, farrowing, and weaning

TRT2	Control									OxBC
T3	Gestation			Farrowing			Weaning			Gestation			Farrowing			Weaning				P-value5
P4	0	1	2+	0	1	2+	0	1	2+	0	1	2+	0	1	2+	0	1	2+	SEM	TRT	T	TRT × T
IgG	134.4	233.6	254.4	101.6	35.2	61.2	115.1	94.9	99.2	49.7	79.6	113.2	86.0	53.6	72.4	86.7	82.1	184.3	45.13	***	**	**
IgM	54.4	47.1	48.9	25.0	22.2	29.8	11.5	13.8	13.5	18.2	17.2	20.7	29.8	27.6	26.2	11.6	17.1	13.9	6.38	**	*	*

¹Samples were diluted to 1:500,000 and reported as ng/mL.

²TRT, dietary treatment.

³ T, time.

⁴ P, parity.

⁵*P < 0.001, **P < 0.05, ***P ≤ 0.10, **** P > 0.05. No farrowing room by treatment interaction was found. There was no significance or tendencies found for parity, dietary treatment × parity, parity × time, or dietary treatment × parity × time.

Colostrum and milk antibodies were not different (P ≥ 0.18) between dietary treatments (Table 4). However, there was a significant (P ≤ 0.04) two-factor interaction between dietary treatments and parity for IgG in the colostrum and both IgG and IgA in the milk. No significant differences among the treatments and paritites were detected in contrast statements with a Tukey adjustment. However, in the colostrum samples parity 1 sows numerically have the lowest IgG in the control group but numerically are the highest in the OxBC treatment. Milk samples from the first parity sows in the control group numerically have the highest IgG and IgA, and the gilts receiving OxBC numerically have the highest IgG where parity 2+ numerically have the highest IgA concentrations. The composition of colostrum and milk did not differ (P ≥ 0.22) between dietary treatments (Table 5). Differences (P < 0.05) were observed across parities for milk fat (parity 1 had higher milk fat than parity 2+), total solids (parity 1 had higher milk fat than parity 2+), true protein (parities 1 and 2+ had lower true protein than gilts), and urea (gilts had higher urea than parity 2+). Furthermore, there was a significant treatment by parity interaction (P < 0.05) for milk total protein. A farrowing room by treatment interaction (P < 0.05) was noted for milk total protein.

Table 4.

Dietary effects of oxidized beta-carotene (OxBC) on sow colostrum and milk immunoglobulins (IgA and IgG)

Diet	Control			OxBC				P-value1
Parity	0	1	2+	0	1	2+	SEM	TRT	Parity	TRT × Parity
Colostrum2
IgG, ng/mL	21.01	19.38	30.38	27.64	34.19	22.89	6.305	0.2516	0.8833	0.0128
IgA, ng/mL	222.21	194.28	300.33	225.19	241.16	241.03	64.462	0.9395	0.3236	0.3824
Milk3
IgG, ng/mL	17.44	27.49	26.04	39.64	14.28	33.84	6.388	0.1786	0.0828	0.0058
IgA, ng/mL	444.45	694.35	554.83	602.06	561.40	688.70	84.821	0.3371	0.2951	0.0384

¹There were no farrowing room by treatment interactions; TRT, dietary treatment.

²Colostrum was diluted 1:1,000,000 for IgG analysis and 1:20,000 for IgA analysis.

³Milk was diluted 1:5,000 for IgG analysis and 1:2,000 for IgA analysis.

Table 5.

Dietary effects of oxidized beta-carotene (OxBC) on sow colostrum and milk composition

Diet	Control			OxBC				P-value1
Parity	0	1	2+	0	1	2+	SEM	TRT	Parity	TRT × Parity	TRT × Room
Colostrum
Fat, %	4.05	5.19	4.81	5.35	5.59	5.15	1.033	0.3577	0.7130	0.8132	0.9616
Lactose, %	2.54	3.38	3.56	3.45	3.62	3.45	1.194	0.6828	0.8365	0.8357	0.9062
Total solids, %	23.21	27.70	26.51	28.08	26.82	30.76	2.989	0.2011	0.3489	0.3415	0.9565
True protein, %	15.91	14.43	14.87	14.75	13.20	16.13	1.105	0.6347	0.0458	0.0966	0.8730
Urea, %	66.01	59.21	58.10	57.75	56.83	59.28	4.702	0.3498	0.5935	0.3676	0.5523
Milk
Fat, %	8.21	9.60	7.80	9.49	9.56	7.53	0.980	0.5892	0.0019	0.5421	0.8001
Lactose, %	5.33	5.31	5.46	5.10	5.34	5.38	0.186	0.4182	0.2435	0.6872	0.7789
Total solids, %	18.11	20.29	17.78	19.64	19.03	18.05	1.272	0.8195	0.0441	0.3589	0.2938
True protein, %	4.53	4.50	4.16	5.10	4.27	4.27	0.200	0.2153	0.0003	0.0463	0.0235
Urea, %	52.61	53.15	50.15	60.02	55.36	48.29	3.846	0.2757	0.0081	0.2021	0.9652

¹TRT, dietary treatment; Room, farrowing room.

There were no differences (P < 0.05 in sow weight across dietary treatments (Table 6). There were no significant differences (P > 0.15) in piglet growth performance, sow reproductive performance, and WEI observed between dietary treatments (Table 6). Additionally, there were no differences (P > 0.05) in sow feed intake (Table 6). There were significant (P < 0.05) parity effects noted where 2+ sows had a higher number of still borns, piglet losses, individual piglet weight at weaning, and feed intake than gilts (Table 6). No differences (P > 0.36) were found between subsequent litters after sows were returned to the common control diet (Table 6). There was a difference (P < 0.05) between sow parities for the number of piglets born alive in the subsequent litters. Data collected from the second group of piglets followed through the nursery provides limited insight into maternal nutrition due to the limitations of the nursery rooms at the farm, which allowed for a statistical N =1 for each diet the piglet had received before weaning.

Table 6.

Effect of supplementation of oxidized beta-carotene (OxBC) on sow reproductive and litter growth performance from farrowing to weaning¹

Diet	Control			OxBC				P-value2
Parity	0	1	2+	0	1	2+	SEM	TRT	Parity	TRT × Parity
Sow body WT
Gestation WT, kg	150.92	156.83	170.27	146.29	154.58	175.38	6.038	0.8034	<0.01	0.3970
Farrowing WT, kg	163.37	164.96	182.51	168.49	165.06	181.16	7.682	0.8604	<0.01	0.8554
Weaning WT, kg	137.72	144.15	168.83	137.14	152.85	169.82	5.129	0.3337	<0.01	0.5997
Sow feed intake
Daily, kg	5.17	6.48	6.87	4.73	6.79	6.59	0.305	0.8344	<0.0001	0.3390
Total Feed, kg	108.47	133.31	145.11	99.99	140.62	138.74	5.959	0.8131	<0.0001	0.3536
Farrowing records
Number piglets born alive	13.89	12.27	14.30	13.55	14.63	14.49	1.228	0.3429	0.5222	0.3889
Litter live born WT3, kg	19.35	18.88	19.72	19.47	22.16	19.87	1.302	0.2428	0.7051	0.3996
Individual live born WT, kg	1.41	1.56	1.41	1.48	1.55	1.44	0.090	0.6520	0.1501	0.8583
Number of still born	0.93	0.68	1.75	0.53	0.16	1.36	0.477	0.1588	0.0028	0.9830
Individual still born WT, kg	1.24	1.33	1.06	1.09	1.10	1.17	0.178	0.5010	0.8278	0.4457
Number of mummies	1.49	0.22	0.73	0.36	1.37	0.37	0.460	0.7044	0.4618	0.0149
Subsequent total born4
Live Born	10.6	14.7	14.4	12.2	13.4	14.0	1.51	0.9804	0.0430	0.5420
Still Born	1.6	1.8	2.6	1.0	0.8	2.2	1.21	0.3677	0.1768	0.9363
Mummy	1.7	1.0	1.2	1.3	1.5	1.3	0.47	0.7189	0.6770	0.5538
Post cross-foster
Litter size	13.61	13.61	12.71	12.84	13.87	13.19	0.599	0.9785	0.2252	0.2730
Adjusted litter start WT, kg	18.93	18.20	17.73	17.72	20.00	18.08	1.269	0.6976	0.4665	0.3503
Piglet losses5	1.28	0.61	2.39	1.04	1.37	2.59	0.547	0.4789	<0.0001	0.5489
Weaning
Number of pigs weaned	12.26	12.87	10.71	11.78	12.51	10.66	0.572	0.4171	<0.0001	0.8306
Litter wean WT, kg	73.65	83.38	70.91	70.86	80.76	69.81	4.540	0.4508	0.0040	0.9526
Individual pig WT, kg	6.02	6.53	6.62	6.03	6.44	6.67	0.342	0.9535	0.0251	0.9693
Litter gain, kg	56.58	63.98	55.09	53.62	61.88	54.52	3.690	0.4213	0.0130	0.8810
Daily litter gain, kg	2.70	3.11	2.62	2.53	3.03	2.60	0.187	0.4366	0.0043	0.8349
Daily pig gain, kg	0.21	0.24	0.24	0.21	0.24	0.23	0.105	0.6500	0.0365	0.9932
Wean-to-estrus interval, days	4.86	4.40	4.68	4.88	3.92	4.48	0.395	0.4088	0.1281	0.7667

¹Sow weights (WT) were estimated using the equation provided in Groesbeck et al. (2002)

²TRT, dietary treatment

³WT, body weight

⁴Records for sows above are from sows that were not culled by the farm after trial.

⁵This number is representative of piglets that died preweaning.

Average daily gain for each treatment was 0.41 kg/day for the piglets from control sows and 0.42 kg/d for the piglets from sows receiving the OxBC dietary treatment. As a percentage, roughly the same number of sows were kept for rebreeding from each treatment (Table 7).

Table 7.

Rebreed status of sows kept in herd¹ to assess dietary effects of oxidized beta-carotene (OxBC) after returning to a common diet at weaning

Diet	Control				OxBC
Parity	0	1	2+	Total	0	1	2+	Total
Percent, %
No heat	2.74	2.74	4.11	9.59	6.58	2.63	3.95	13.16
Pregnant	19.18	20.55	42.47	82.19	18.42	13.16	40.79	72.37
Not Pregnant	5.48	0	2.74	8.22	5.26	1.32	7.89	14.47
Total	27.4	23.29	49.23	100	30.26	17.11	52.63	100

¹Sows were culled from the herd as determined by the farm protocol.

The plasma vitamin A status of both the sows and piglets did not differ (P > 0.05) between dietary treatments (Table 8). There were differences (P < 0.05) across time points and a two-factor interaction (P < 0.05) between time (birth 0.04 ppm vs. weaning 0.16 ppm) and parity for plasma vitamin A status of the pigs in which piglets from higher ­parity sows had higher vitamin A levels. It should be noted that due to the nature of the blood from the umbilicord, piglet birth blood is a mixture of both sow and piglet blood. Vitamin A status of the piglet liver was not different (P = 0.82) between dietary treatments, but there was a difference (P = 0.04) across time (birth 5.5 ppm vs. weaning 23.9 ppm) and between time and dietary treatment (Table 9).

Table 8.

Effects of dietary effects of supplemental oxidized beta-carotene (OxBC) on plasma vitamin A as retinol (ppm) status of sampled sows and pigs

TRT1	Control									OxBC
T2	Gestation			Farrowing			Weaning			Gestation			Farrowing			Weaning				P-value4
P3	0	1	2+	0	1	2+	0	1	2+	0	1	2+	0	1	2+	0	1	2+	SEM	P	T	P × T
Sow	0.17	0.14	0.18	0.17	0.18	0.18	0.18	0.19	0.21	0.17	0.16	0.18	0.18	0.17	0.16	0.15	0.23	0.21	0.020	***	*	*
Piglet5	–	–	–	0.03	0.04	0.05	0.15	0.18	0.18	–	–	–	0.04	0.06	0.04	0.13	0.17	0.18	0.010	*	*	**

¹TRT, dietary treatments.

²T, time.

³P, parity.

⁴*P < 0.001, **P < 0.05, ***P > 0.05. No tendencies were found for plasma vitamin A. No farrowing room by treatment interaction was found. There were no significant differences or tendencies found for dietary treatment, dietary treatment × parity, dietary treatment × time, and dietary treatment × parity × time.

⁵Piglet blood collected from the umbilical cord at birth is a mixture of both sow and piglet blood due to the nature of the umbilical cord blood.

Table 9.

Dietary effect of maternal nutrition supplemented with oxidized beta-carotene (OxBC) on piglet liver vitamin A (ppm) status at birth and weaning

Diet	Control		OxBC			P-value 1
Time	Birth	Weaning	Birth	Weaning	SEM	TRT	Time	TRT × Time
Piglet liver2	5	24	6	24	0.90	0.5575	<0.0001	0.4377

¹TRT, dietary treatment.

²Liver vitamin A was measured by converting all retinyl palmitate to retinol, and measuring the subsequent retinol level.

Discussion

Oxidized beta-carotene has been reported to support the immune system and piglet growth pre-weaning (Jun et al., 2021). Improving the immune status of sows and optimizing piglet growth prior to weaning is important. Sow mortality has increased over the past years from 10.98% in 2016 to 13.56% in 2021 (Eckberg, 2022) along with prewean mortality from 11.9% in 2007 to 13.1% in 2016 (Koketsu et al., 2021), thus any mitigation strategies are pertinent to helping increase livability.

Although previous work has seen immunological benefits in feeding OxC-beta, there are reasons why there were no changes in the measured immune parameters. One reason could be the number of animals evaluated in the different studies. An increase in the number of animals sampled will help eliminate the potential for error and maximize accuracy (Osborne and Costello, 2004). Additionally, there could be herd-to-herd variation from where the studies are conducted. Documented in cows, differences in genetics, nutrition, and general herd management, can affect herd performance (Poppe et al., 2021).

Furthermore, the evaluation of other innate immune system components to assess the effects of dietary treatment could be explored, including neutrophils and lymphocytes. The receptors measured in this study included a CD4 antigen that had the potential to bind four populations of resting T lymphocytes including CD4+/CD8-, CD4-/CD8+, CD4-/CD8-, and CD4+/CD8+ cells. These T lymphocytes are responsible for recognizing antigens that come from either pathogens, tumors, or the environment (Kumar et al., 2018). Additionally, CD8 alpha can bind to cytotoxic T cells, thymocytes, and porcine natural killer cells. Cytotoxic T cells and natural killer cells are both effector cells in the immune system, that can both attack harmful cells. Cytotoxic T cells, part of the adaptive immune system, will recognize a foreign antigen before destroying the harmful cell (Murphey et al., 2008). Natural killer cells, part of the innate immune system, will have nonspecific recognition of tumor cells, or cells with no major histocompatibility complex 1, which is found on all nucleated cells (Murphey et al., 2008). Thymocytes are immune cells that are in the thymus and are destined to become a T cell (Murphey et al., 2008). Finally, CD335 is known to bind to natural killer cells, but it is not expressed on all porcine natural killer cells. Thus, there is potential to further investigate specific immune cell populations to assess more definitive effects of oxidized beta-carotene on the cells in the immune system. Both T cells and natural killer cells are important to help recognize, specific and nonspecific, antigens in the body and provide a defense mechanism to get rid of them. While the phagocytic activity of Kupffer cells was assessed, a similar assay could be done with circulating macrophages.

A trend was observed for a decrease in milk IgG over time, concomitant with an increase in milk IgA (Curtis and Bourne, 1971). Colostrum composition in this study is similar to previously reported compositions (Klobasa and Butler, 1987; Jackson et al., 1995; Jun et al., 2021). Milk composition is also similar (Klobasa and Butler, 1987; Jackson et al., 1995), except for an observed higher fat when compared to Jun et al. (2021).

Differences between parity performance are expected as P0 and P1 sows have different reproductive performance than more mature sows (Koketsu et al., 2017). Some differences, such as a higher still born rate were not expected. This may have been influenced by including sows in this study beyond parity 5, as it has been noted that sow reproductive performance peaks at parity 5 and then starts to decrease (Koketsu et al., 2017). Although overall, there was no difference in sow reproductive performance between the dietary treatments.

The plasma vitamin A status of the sows at farrowing was not consistent with other published literature or trends (Jun et al., 2021; Hinson et al., 2022), as a decrease in vitamin A levels was not observed. One reason for the decline in vitamin A not being observed could be that the blood sample taken from the sows in previous studies may not have been taken directly at the time of farrowing. For example, Hinson et al. (2022) collected blood samples within 24 h of farrowing, while the study herein collected blood samples as sows were actively farrowing. It is possible that after farrowing occurs, vitamin A is mobilized to reproductive tissues to help with inflammation or to the mammary gland to be transferred to the piglet via colostrum. Except for sows during farrowing, the vitamin A plasma levels for the piglet and sow and piglet liver vitamin A are consistent with other recently published literature (Hinson et al., 2022). It should be noted that since blood was taken from the umbilical cord of the piglet, the collected blood was a mixture of both sow and piglet blood due to the nature of the blood in the umbilical cord. Plasma vitamin A is highly regulated (Tanumihardjo, 2011). Thus, vitamin A levels should be assessed in the liver of animals for reliable vitamin A assessment. Previous work (Valentine and Tanumihardjo, 2005) demonstrated that piglet liver vitamin A levels are increased when the sow received a megadose of vitamin A. The study herein did not observe a difference in hepatic vitamin A. Thus, this information confirms that OxBC is a nonvitamin A compound and does not influence vitamin A status in pigs following consumption.

The data herein suggest that the supplementation of OxC-Beta at 80 ppm from day 60 of gestation through lactation does not affect the reproductive performance of sows during the current or subsequent litter, piglet growth performance, vitamin A status of the piglet or sow, piglet immune status, and antibodies or composition in colostrum and milk. The supplementation of OxC-Beta did decrease IgM and tended to decrease IgG in sow plasma.

Glossary

Abbreviations

CD

cluster of differentiation

FTU

phytase units

GE

gross energy

Ig

immunoglobulin

OxC-Beta

oxidized beta-carotene

P

parity

PBS

phosphate buffer saline

SID

standardized ileal digestible

T

time

TRT

dietary treatment

WEI

wean to estrus interval

WT

body weight

Conflict of Interest Statement

Funding was provided by the company which provided the product for evaluation.