Impact of arginine supplementation on serum prolactin and mRNA abundance of amino acid transporter genes in mammary tissue of lactating sows

Abstract

This study was conducted to test the hypothesis that supplemental dietary Arg to late-pregnant and lactating sows increases serum prolactin concentrations and mRNA abundance of SLC7A1, SLC7A2, and SLC6A14 in mammary parenchymal tissue. From day 108 of gestation and until day 21 of lactation, sows were fed a diet either supplemented with 0.10 g of l-Arg/kg body weight (BW) per day (n = 10, ARG) or 0.34 g of l-Glu/kg BW per day (n = 10, control). Litters were standardized to 10 piglets on day 1 of lactation and piglets were weighed on days 1, 7, 14, and 21 of lactation. Sow BW was recorded on day 108 of gestation and days 1, 10, and 21 of lactation. Lactation sow feed intake was recorded daily. Mammary parenchymal tissue was biopsied on day 5 of lactation to measure mRNA abundance SLC7A1, SLC7A2, and SLC6A14. On days 4 and 18 of lactation, blood samples were collected from sows at 2, 4, and 6 hr postfeeding to measure serum prolactin concentrations. Milk samples were collected on days 4, 10, and 18 of lactation to measure fat, lactose, urea N, and true protein concentrations. Sow BW, backfat, and feed intake over all sampling days did not differ between treatments. Piglet BW on d 1 tended to be greater for the ARG treatment than the control treatment (P = 0.12). Sow milk yield and composition (fat, protein, lactose, and urea N) and mammary mRNA abundance of candidate genes did not differ between the ARG and the control group. Compared to controls, serum prolactin concentrations tended to be greater (P = 0.08) in ARG sows on day 4 of lactation, and did not differ on day 18. Current findings show a potential beneficial effect of dietary supplementation with Arg to late-pregnant multiparous sows on BW of their piglets on day 1. Dietary Arg supplementation at a rate of 0.10 g/kg BW during late pregnancy and lactation tended to increase serum prolactin concentrations with no increase in mammary transcript abundance of SLC7A1, SLC7A2, and SLC6A14 in early lactation.

Introduction

Arginine is a semi-essential amino acid (AA) in growing animals and is non-essential in adults, but its dietary requirement for lactation has not been empirically determined. The factorial estimate of Arg requirement for lactation (NRC, 2012) is uncertain because its maximum biological efficiency value for lactation remains to be validated (Zhang et al., 2019). Thus, it is possible that Arg requirement for lactation may be underestimated. Arginine is well known as an important functional AA involved in the regulation of numerous physiological pathways (Wu, 2009). Among those, 2 are particularly relevant to lactation, including regulation of blood flow via nitric oxide synthesis (Cieslar et al., 2014) and stimulation of prolactin secretion across multiple animal models (Thompson et al., 2017; Zhu et al., 2017). In lactating sows, prolactin is essential for milk production (Farmer et al., 1998). Therefore, Arg has gained interest as a potential dietary supplement in lactating sows. Pérez-Laspiur et al. (2006) supplemented sow diets with Arg during lactation and noted an improvement in sow body weight (BW) retention potentially related to changes in both insulin and somatotropin concentrations. Mateo et al. (2008) reported improved weight gain of suckling piglets from sows fed supplementary Arg. Prolactin was shown to increase SNAT2 protein abundance during lactation in rat mammary glands (Velázquez-Villegas et al., 2015) and to regulate LAT1 expression and activity in mammary epithelial cells of dairy cows (Zhou et al., 2020). The SNAT2 protein is part of system A AA transporter and mediates Na⁺-coupled cellular uptake of small neutral AA, and LAT1 mediates the uptake of large neutral AA. Supplemental Arg may improve litter gain via mediation of prolactin secretion and transcription upregulation of genes encoding for cationic AA transporters, in particular that of Lys, the first limiting AA in lactating sow diets. In a recent study, however, mammary uptake of AA and energy metabolites in lactating sows were unaffected by dietary Arg supplementation (Krogh et al., 2017). Because of the intensive nature of that study, only 4 sows per treatment were used and prolactin concentrations were not measured. The hypothesis of the current study was that supplemental dietary l-Arg at a rate of 0.10 g per kg BW to late-pregnant and lactating sows, increases serum prolactin concentrations and mRNA abundance of SLC7A1, SLC7A2, and SLC6A14 in mammary parenchymal tissue.

Materials and Methods

All animal procedures in this study were performed with the approval of the Institutional Animal Care and Use Committee at Michigan State University (AUF number 05/02-105-00).

Animals, experimental treatment, and housing

Twenty multiparous Yorkshire sows were bred with semen from PIC boars. On day 108 of gestation, 20 multiparous sows were stratified for BW and parity to receive 1 of 2 dietary treatments, with 10 sows per treatment: (1) l-Arg (ARG, 0.10 g/kg BW) or (2) l-Glu (control, 0.34 g/kg BW) (Ajinomoto Animal Nutrition North America, Chicago, IL). Both l-Arg and l-Glu were top-dressed daily. Sow BW on day 108 of gestation and on day 1 of lactation was used to adjust the daily supplementation dose between day 108 of gestation to day 2 of lactation, and days 3 to 21 of lactation, respectively. Glutamic acid was used to ensure that the control and ARG sows received iso-nitrogenous diets. Arginine and Glu were supplemented on a BW basis to maintain a constant AA intake independent of feeding level. The rate of Arg supplementation to a 250-kg sow in the ARG group corresponded to a dietary concentration of 1% with a feed level of 2.5 kg/d (i.e., 25 g/sow/d), which is within the range used in previous studies (Mateo et al., 2008; Zhu et al., 2017). Sows were housed in farrowing crates with ambient temperature maintained at 20 °C and lights on for 16 hr daily. On day 1 of lactation (day 0 being farrowing day), litters were standardized to 10 piglets by removing surplus piglets to litters with <10 piglets within the same treatment groups. Newborn piglets from nonexperimental sows were used in the case of an insufficient no of surplus piglets within treatment groups.

Performance data and samplings

Sow BW was recorded on day 108 of gestation and on days 1, 10, and 21 of lactation. The average of 3 measurements of sow backfat thickness was recorded on days 1 and 21 of lactation using an ultrasound equipment fitted with a 3.5-MHz probe and a 12.6-cm probe head (ALOKA SSD-500V; Hitachi Aloka Medical, Ltd., Tokyo). Measurements were obtained 5 cm lateral to the spine and were centered over the last posterior rib. Piglets were weighed individually on days 1, 7, 14, and 21 of lactation.

A total of 50 mL of milk was manually obtained after emptying 2 to 3 abdominal mammary glands, subsequent to the intramuscular administration of 1 mL of oxytocin (20 IU/mL; VetTek, Blue Springs, MO) on days 4, 10, and 18 of lactation. Pigs were separated from their dam for ~60 min before oxytocin was injected. On days 4 and 18 of lactation, blood samples (9 mL) were collected at 2, 4, and 6 hr postfeeding of the morning meal (day 4: 1.61 kg/meal; day 18: 2.63 kg/meal). Samples were obtained from an ear vein using a 21-gauge butterfly needle connected to a 20-cm long tube equipped with a luer slip port. Blood was gently drawn using a syringe and transferred into vacutainers (RST Tube with Thrombin-Based Clot Activator and Polymer Gel; BD Medical Supplies, Franklin Lakes, NJ). Blood was centrifuged at 1,500 × g for 20 min at 4 °C and serum was stored at −20 °C until analysis.

Mammary parenchymal tissue was biopsied on day 5 of lactation from the first or second thoracic mammary gland, as described by Kirkwood et al. (2007). Mammary parenchymal tissue (~1.0 g) was obtained 2 hr after feeding and immediately after completion of a nursing bout. Tissue was rapidly frozen in liquid N and stored at −80 °C until analysis. During collection of mammary tissue, piglets were isolated in an adjacent pen equipped with a heating lamp. Approximately 3 hr after biopsy, piglets were returned to the sow.

Diets and feeding

A corn soybean meal-based lactation diet was formulated to supply optimal energy and macronutrients levels according to the nutrient requirements of lactating sows (NRC, 2012). Calculations were based on a lactation length of 21 d, a litter size of 10 piglets with an average daily gain of 250 g/d, a sow BW loss during lactation of 10 kg and a predicted average daily feed intake of 6.4 kg. The lactation diet was formulated to contain 9.1 g SID Lys/kg DM and 9.1 g SID Arg/kg DM. Dietary content and nutritional composition are shown in Tables 1 and 2, respectively. From day 108 of gestation until day 2 of lactation, sows were fed 26.9 MJ of NE/d (equivalent to 2.5 kg/d) of the lactation diet. After parturition, feed allowance was gradually increased to 67.9, 81.9, and 86.2 MJ NE/d (equivalent to 6.3, 7.6, and 8.0 kg/d, respectively) on days 7, 14, and 21 of lactation, respectively. The daily feed allowance was divided into 3 equally sized meals fed in 8 hr-intervals throughout the experiment. The daily supplementation of Arg (0.10 g/kg BW) and Glu (0.34 g/kg BW) was divided into 3 portions each thoroughly mixed into ~100 g of feed and fed just prior to rest of the meal. Sow feed intake was recorded on a daily basis throughout the experiment and diet was sampled weekly and pooled for nutrient analysis. Piglets had no access to creep feed and water was freely available to sows and piglets.

Table 1.

Ingredient composition of the lactation diet fed to sows from day 108 of gestation to day 21 of lactation

	Lactation diet
Ingredients, g/kg
Corn, yellow dent	692
Soybean meal, high protein expelled	175
Sugar food products (SFP)1	50
Soybean hulls	15
Choice white grease	27
Limestone, ground	13.0
Mono calcium phosphate	13.0
Sodium chloride	4.70
Vitamin premix2	5.00
Mineral premix3	1.25
Se 2004	0.70
L-Lys-HCL	2.60
DL-met	0.10
L-Thr	0.70
L-Trp	0.10

¹Candy production by-product (International Ingredient Corporation, St. Louis, MO), content: 11.9 MJ NE, 0.5 g fermentable fiber/kg and 10 g CP/kg.

²Supplied per kilogram of diet: vitamin A: 6,000 IU; vitamin D3: 600 IU; vitamin E: 40 IU; vitamin K: 2 mg; vitamin B12: 35 µg; riboflavin: 8 mg; d-panthotenic acid: 20 mg; niacin: 30 mg; biotin: 200 µg; folic acid: 1.50 mg; choline (as choline chloride): 500 mg; vitamin B6 (as pyridoxine HCl): 4.5 mg; vitamin E (as dl-tocopherol acetate): 20 IU; chromium (as Cr picolinate): 180 µg; l-carnitine: 45 mg.

³ Supplied per kilogram of diet: Fe: 1,280 mg as iron carbonate; Zn: 518 mg as sinc oxide; Mn: 72 mg as manganese oxide; Cu: 39 mg as copper chloride; I: 1.25 mg as ethylenediamine dihydroiodide.

⁴Provided 0.5 mg selenium per kilogram of diet as Na₂Se.

Table 2.

Calculated and analyzed energy and nutrient composition of the lactation diet fed to sows from day 108 of gestation to day 21 of lactation

	Lactation diet
Calculated composition
Net energy (NE), MJ/kg DM	12.3
Total crude protein, g/kg DM	166
SID crude protein, g/kg DM	137
STTD P, g/kg DM	4.2
Fermentable fiber, g/kg DM	111
Total P, g/kg DM	6.7
C, g/kg DM	432
Ca, g/kg DM	8.4
Na, g/kg DM	2.5
Cl, g/kg DM	5.4
SID amino acid content, g/kg DM
Lys	9.1
Met	2.5
Thr	5.7
Ile	5.7
Leu	13.0
Val	6.4
Phe	6.8
His	3.9
Arg	9.1
Analyzed composition
DM, g/kg feed	877
Gross energy, MJ/kg DM	18.44
Ash, g/kg DM	72
Starch, g/kg DM	512
Nonstarch polysaccharides, g/kg DM	100
Lignin, g/kg DM	14
Dietary fiber, g/kg DM	114
Fat, g/kg DM	47
Crude protein, g/kg DM	181
Amino acids, g/kg DM
Lys	11.5
Met	3.2
Thr	8.1
Ile	8.3
Leu	18.0
Val	9.6
Phe	9.5
His	5.1
Arg	11.8

Chemical analysis

Dietary dry matter (DM) content was measured by oven drying to a constant weight (~20 hr) at 103 °C, and ash was analyzed according to AOAC (2005; Method 942.05). Nitrogen was measured by the Dumas method (Hansen, 1989) and CP was calculated as N × 6.25. Gross energy was determined with an adiabatic bomb calorimeter (IKAC 400, IKA Werke, Janke & Kunkel, Staufen, Germany). Starch and nonstarch polysaccharides were analyzed as described by Bach Knudsen (1997), and Klason lignin was measured as the sulfuric acid-insoluble residue (Theander and Aman, 1979). Fat was extracted with a monophasic mixture of chloroform, methanol, and water after hydrochloric acid hydrolysis as described by Bligh and Dyer (1959). Dietary samples were hydrolyzed for 23 hr at 110 °C with (Met) or without (Lys, Met, Thr, Ile, Leu, Val, Phe, His, and Arg) oxidation using performic acid, and AA were separated by means of ion exchange chromatography and quantified by photometric detection after the ninhydrin reaction (European Commission, 1998).

Milk fat, lactose, urea nitrogen, and protein concentrations were assayed using infrared spectroscopy (Multispec M, Berwind Instrument Ltd., York, UK). Serum prolactin concentrations were determined according to a previously described RIA (Robert et al., 1989). The radioinert prolactin and the primary antibody to prolactin were purchased from A. F. Parlow (U.S. National Hormone and Pituitary Program, National Institute of Diabetes and Digestive and Kidney Diseases, Torrence, CA). Assay validation was performed using pooled serum from lactating sows. Parallelism was 98.3% and average mass recovery was 99.9%. Sensitivity of the assay was 1.5 ng/mL. The intra- and interassay CV were 4.68% and 2.79%, respectively.

Mammary parenchymal tissue samples were homogenized and the ribonucleic acid was extracted using a commercial available kit (mirVana miRNA Isolation Kit, Ambion, Life Technologies Inc., Grand Island, NY). An Agilent Bioanalyzer 2100 with RNA 6000 Nano Labchip (Agilent Technologies, Palo Alto, CA) was used to determine RNA purity and quantity. Complementary DNA was synthesized using a commercial available reverse transcriptase kit (High capacity RNA-to-cDNA kit, Applied Biosystems, Life Technologies Inc., Grand Island, NY). The complementary DNA concentration was quantified by spectrophotometry (NanoDrop 1000; Thermo Scientific, Waltham, MA), diluted to a working stock of 10 ng/μL and stored at −20 °C. Reverse transcription quantitative PCR was run on an ABI StepOne Plus Real-Time PCR System (Applied Biosystems) including 2 initial steps (50 °C for 2 min and 95 °C for 20 s). The following 40 cycles consisted of 95 °C for 1 s followed by 60 °C for 20 s. The porcine-specific genes VAPB (vesicle-associated membrane protein-associated protein B/C) and RSP23 (Ribosomal protein 23) were used as reference genes (Applied Biosystems). In addition, exon 2 of the human equivalent of the MRPL39 (Mitochondrial ribosomal protein L39) gene was used to construct a custom TaqMan assay and used as the third reference gene. Three commercial available Taq-Man assays for porcine-specific genes were studied (Applied Biosystems). The genes of interest were SLC6A14, SLC7A1, and SLC7A2, coding for the AA transporter proteins ATB^0,+, CAT-1, and CAT-2b, respectively.

Calculations and statistical analyses

Total nonstarch polysaccharides were based on the determination of sugar residues, and dietary fiber was quantified as total nonstarch polysaccharides plus lignin. Sow milk yield was calculated based on the average litter size and litter gain for each week in lactation using the prediction model described by Hansen et al. (2012).

The differences in transcript abundance between each gene of interest and the reference genes were calculated (ΔC_t) in order to normalize the data. The ΔC_t values were calculated as the difference between the threshold cycle (C_t) of the gene of interest and the average C_t of the reference genes. Statistical analysis was conducted on the ΔC_t values. Then, the ΔΔC_t values for the genes of interest were calculated as: average ΔC_t value of ARG sows – LSMean ΔC_t value of control sows. Finally, the relative expression of the AA transporter genes in mammary tissue of ARG sows relative to that of control sows was calculated as 2−^ΔΔCt. Consequently, values lower than 1 indicate a lower expression in ARG sows than in control sows, while the opposite is true for values above 1.

Statistical analyses were conducted with the Mixed procedure of SAS (SAS Inst. Inc., Cary, NC). The statistical model for sow performance variables (Table 3) and mRNA abundances (Figure 1) included treatment (ARG, control) as a fixed effect and farrowing room (1, 2) as a random effect. The statistical model for milk composition and performance characteristics with more than one observation per sow (Table 4) included the fixed effects of treatment (ARG, control), week in lactation (week 1, 2, 3, or week −1, 1, 2, 3 for sow BW and feed intake) and their interactions. The farrowing room was included as a random effect. The repeated observations taken from each sow or litter was accounted for by including a first-order autoregressive covariance structure using the repeated statement and specifying the sow as the subject to repeated measures. The statistical model for serum prolactin concentrations (Figure 2) were analyzed including treatment (ARG, control), day of lactation (d4 and d18), sampling time (2, 4, and 6 hr postfeeding) and their interaction as fixed effects. The farrowing room and day of lactation within each sow were included as random effects. The repeated samples taken from each sow within sampling day was accounted for by including a first-order autoregressive covariance structure using the repeated statement and specifying the sow within day of lactation as the subject to repeated measures. Due to a tendency for a treatment × day interaction and a treatment × time post-feeding interaction, separate ANOVA analyses were done on prolactin concentrations on days 4 and 18. In the separate analyses, a treatment effect was observed on day 4, but not on day 18, and no treatment × time postfeeding interactions were observed. Thus, the final ANOVA analyses on separate days included treatment (ARG, control) as a fixed effect and time postfeeding and farrowing room as random variables. One control sow was excluded from the experiment on day 6 of lactation due to low feed intake (<2.0 kg/d between day 110 of gestation and day 5 of lactation) and 1 ARG sow was excluded due to hypersensitivity to anesthesia. Differences between treatments were declared significant at P ≤ 0.05 and a tendency at P ≤ 0.15.

Table 3.

Sow and litter performance for sows supplemented with 0.10 g arginine/kg BW/d (ARG) or 0.34 g glutamic acid/kg BW/d (control) from day 108 of gestation until day 21 of lactation

	Control	ARG	SEM	P-value
n	9	9
Parity number	5.2	5.3	0.8	0.92
Gestation length, d	116	116	0.4	0.84
Live born	9.9	9.6	1.4	0.86
Stillborn	1.0	0.8	0.31	0.62
Piglet BW, day 1, kg1	1.49	1.67	0.08	0.12
Piglet BW, day 1, kg2	1.50	1.67	0.08	0.14
Litter size, day 21	9.1	9.6	0.28	0.27

¹Prior to litter standardization.

²After litter standardization

Figure 1.

The mRNA abundances of SLC7A1, SLC7A2, and SLC6A14 on day 4 of lactation presented as fold changes in ARG sows (n = 9) (0.10 g l-Arg/kg BW/d) relative to control sows (n = 9) (0.34 g l-Glu/kg BW/d). Fold changes are presented on a logarithmic scale. Error bars indicate upper and lower 95% confidence intervals.

Table 4.

Lactation performance and milk composition of sows supplemented with 0.10 g Arg/kg BW/d (ARG) or 0.34 g Glu/kg BW/d (control) from day 108 of gestation until day 21 of lactation

	Trt1			Week relative to parturition					P-value
	control	ARG	SEM	−1	1	2	3	SEM	Trt1	Time	Trt × Time
N	9	9
Sow BW, kg2	245	237	7	246a	238b	239b	241b	5	0.37	<0.001	0.11
Average piglet BW, kg3	3.59	3.94	0.18	—	2.11	3.57	5.61	0.15	0.39	<0.001	0.35
Sow backfat, mm4	15.4	15.8	1.3	—	15.5	—	15.6	0.9	0.80	0.75	0.15
Sow feed intake, kg/d	5.2	5.2	0.1	2.5a	4.4b	6.6c	7.3d	0.1	0.83	<0.001	0.68
Litter ADG, kg	2.26	2.52	0.18	—	1.74a	2.52b	2.88c	0.16	0.48	<0.001	0.28
Piglet ADG, g	234	250	15	—	154a	264a	309a	13	0.63	<0.001	0.20
Milk yield, kg/d	8.9	9.4	0.3	—	6.8a	9.4b	11.1c	0.3	0.17	<0.001	0.17
Milk chemical composition5
Fat, %	7.0	6.6	0.2	—	7.0	6.7	6.8	0.2	0.23	0.46	0.35
Protein, %	4.4	4.3	0.1	—	4.6a	4.3b	4.2b	0.1	0.84	<0.001	0.77
Lactose, %	5.59	5.64	0.04	—	5.39a	5.74b	5.73b	0.04	0.32	<0.001	0.90
Urea N, mg/dL	7.7	7.5	0.5	—	5.3a	8.5b	9.0b	0.5	0.72	<0.001	0.73

¹Trt: effect of treatment (ARG vs. control).

²Values represent the average values of individual BW recorded on day 108 in gestation and on days 1, 10, and 21 of lactation.

³Piglet BW averaged over weeks 1, 2, and 3. The weeks 1, 2, and 3 values were the average BW of days 1 and 7, 7 and 14, and 14 and 21, respectively.

⁴Value represents the averaged value of individual backfat recorded on days 1 and 21 of lactation.

⁵Values represent the mean composition values of individual milk samples collected on days 4, 10, and 18 of lactation.

Figure 2.

Prolactin concentrations (ng/mL) 2, 4, and 6 hr after feeding on (A) day 4 and (B) day 18 of lactation in serum from lactating sows supplemented with 0.10 g l-Arg/kg BW/d (ARG, solid line and filled circle, n = 9) or 0.34 g l-Glu/kg BW/d (control, dashed line and open circles, n = 9) from day 108 of gestation until day 21 of lactation. Serum prolactin decreased between days 4 and 18 of lactation (P < 0.001) and with time after feeding (P < 0.001). † ARG vs. control tends to differ on day 4 overall in lactation (P = 0.08). Data are LSMeans ± SEM.

Results

Analyzed chemical composition of the lactation diet was within the expected range for all nutrients (Table 2). Number of live born, still born, and weaned piglets did not differ between treatments (Table 3). Piglet BW of ARG sows prior to litter standardization tended to be greater (P = 0.12) than piglet BW of control sows (Table 3). Table 4 shows lactation performance and milk composition of sows. Average BW, backfat, and feed intake of sows over all sampling days did not differ between treatments. Litter and individual pig ADG as well as piglet BW over lactation did not differ between treatments. Sow BW decreased between late gestation and the first week of lactation (P < 0.001) and remained stable thereafter. Backfat thickness did not differ between treatments in the 1st and 3rd weeks of lactation.

Milk yield, piglet BW, litter BW gain, and piglet BW gain increased with each week of lactation (P < 0.001). Sow milk yield and composition (fat, protein, lactose, and urea nitrogen) did not differ between treatments. Milk fat concentration was not affected by week of lactation and milk lactose and urea N concentrations increased between weeks 1 and 2 of lactation (P < 0.001) and remained constant thereafter. Milk protein concentration decreased between weeks 1 and 2 of lactation (P < 0.001) and remained stable thereafter. Transcript abundance of SLC6A14, SLC7A1, and SLC7A2 in mammary parenchyma did not differ between treatments (Figure 1).

Serum prolactin concentrations in sows on days 4 and 18 of lactation are shown in Figure 2. Concentrations of prolactin decreased with time after feeding (P < 0.001) and between days 4 and 18 of lactation (P < 0.001). There was a tendency for a treatment × day interaction (P = 0.11) and a treatment × time postfeeding interaction (P = 0.11). Separate analyses on days 4 and 18 showed that the average serum prolactin concentration on day 4 tended to be greater in ARG sows (41.4 ng/mL) than in control sows (35.0 ng/mL; P = 0.08), with no differences observed on day 18 (P = 0.80).

Discussion

The goal of this study was to determine whether Arg supplementation increases prolactin concentration and the transcript abundance of candidate genes known to be important in regulating cationic AA transport in epithelial cells. An iso-nitrogenous supplementation in the 2 treatment groups was applied to ensure that any observed differences were related to the functional effects of Arg rather than to the N supply. Therefore, compared to ARG diet, the control diet contained equivalent moles of N through Glu supplementation. Dietary Glu is greatly metabolized in the enterocytes and consequently functioning as a postabsorptive supply of N with no specific impact on postabsorptive metabolism (Tome, 2018). This study was not designed to assess the impact of Arg supplementation on production performance, consequently, the number of sows per dietary treatment was low. Therefore, tendency for difference between treatments in production results was accepted at an α level ≤0.15, which should be emphasized when interpreting results. Nonetheless, reporting such values are critical to demonstrate that sows in both treatments were healthy and performing as predicted based on NRC (2012) model. Mean BW of piglets throughout lactation were greater by 9% in ARG litters compared with control litters, agreeing with Mateo et al. (2008) who reported 8.3%, 8.7%, and 7.6% increase in piglet BW on days 7, 14, and 21 of lactation, respectively, in sows fed supplemental Arg. In contrast to Mateo et al. (2008) and Zhu et al. (2017), piglet ADG was not affected by supplemental Arg. Thus, it is possible that the enhanced lactation BW of piglets in the current study was linked to their higher BW on day 1 (1.67 vs. 1.49 kg) and not to Arg supplementation per se in lactation, as supported by the lack of difference in ADG between treatments (Table 4). Parenteral administration of Arg to late pregnant ewes increased mammary cell proliferation (Sciascia et al., 2019), therefore colostrum production may have been improved in this study. Supplementary Arg in a study by Krogh et al. (2016), however, did not increase colostrum production. Mateo et al. (2008) saw no effect of supplementary Arg in gestation on piglet BW on the day of farrowing, although BW gain during the colostral period (birth to 24 hr postpartum) was not recorded. Alternatively, Arg in the last week of gestation may increase fetal nutrient availability (Oksbjerg et al., 2019) through its role in angiogenesis (Holanda et al., 2019).

Arginine increased piglet ADG by close to 10% in primiparous sows (Mateo et al., 2008), 6% in sows with an average parity of 2.5 (Zhu et al., 2017) and had no effect in sows with an average parity of 3.0 (Krogh et al. 2016). The average parity in the present study was 5 with Arg supplementation levels (0.5% to 0.3% of diet) similar to that of Krogh et al. (2016) and lower than that of Mateo et al. (2008) (0.5% to 1% of the diet). Arginine supplementation to second and third parity sows did not increase mammary plasma flow and uptake of AA or energy metabolites (Krogh et al., 2017). Bass et al. (2017) also reported that Arg supplementation at 1.0% of the diet fed to a large cohort of sows, either primiparous or multiparous under commercial settings, did not impact lactation performance. Thus, taken together, although Arg requirement may be higher in younger sows, in particular primiparous, the impact of supplementation on milk yield appears to be marginal at best.

Zhu et al. (2017) suggested that dietary supplementation with Arg during lactation could lead to endocrine changes. Stimulation of prolactin secretion by Arg has been demonstrated across multiple animal models including horses (Thompson et al., 2017), cows (Chew et al., 1984), and sows (Pérez Laspiur et al., 2006; Zhu et al., 2017). In Pérez Laspiur et al. (2006), overall concentrations (days 7, 14, and 21 of lactation) were greater in sows fed 1.34% or 1.73% Arg, compared to 0.96% Arg, under thermoneutral conditions. It was apparent that the increase in prolactin in response to Arg feeding was more pronounced on day 7 than in later lactation, with similar concentrations as reported here. In the current study, prolactin concentrations decreased from 2 to 6 hr after feeding, suggesting an impact of Arg, or other nutrients, on prolactin release. For instance, refeeding after a 24-hr fast was shown to increase prolactin concentrations in lactating sows (Rojkittikhun et al., 1993). Similar to Pérez Laspiur et al. (2006), prolactin in Arg-fed sows tended to be greater than control-fed sows on day 4 and not on day 18 of lactation, and notably so at 4 hr postfeeding. Thus, sows appear to respond differently to Arg supplementation in early lactation compared with peak lactation. Arginine intake when expressed per kilogram of feed, declined from 0.5% (4.5% of SID protein intake) on day 4% to 0.3% (2.7% of SID protein intake) on day 18, which could explain in part the difference in response between early and peak lactation. Sows by Pérez Laspiur et al. (2006), however, had increased Arg intake with progression of lactation and reported similar prolactin response as here. When using the dopamine agonist bromocriptine to inhibit prolactin synthesis, Farmer et al. (1998) demonstrated that prolactin is required for galactopoiesis at any week of lactation (Farmer et al., 1998). In addition, piglet BW response to higher prolactin concentrations in lactation is greater in late lactation (day 22) than earlier (day 7; Farmer, C., unpublished data). Recently, it was reported that Arg supplementation decreased the transcript abundance of PRLR (prolactin receptor) in sow mammary tissue (Holanda et al., 2019), and thus the increased prolactin concentration in response to Arg supplementation may simply be a compensatory mechanism. For instance, prolactin binding affinity (Plaut et al., 1989) and abundance of PRLR mRNA (Manjarín et al., 2012) are lower in early lactation and increase with progression of lactation, while prolactin serum concentrations are higher in early lactation and decrease with progression of lactation. Nonetheless, mammary biopsies were obtained in early lactation, when prolactin concentrations tended to be greater in ARG sows. Prolactin is involved in the synthesis of casein, whey acidic protein and lactalbumin (Ben-Jonathan et al., 2006), in part via regulation of nutrient transporter proteins and stimulation of mammary epithelial cell proliferation (Liu et al., 2020). Mammary arteriovenous difference of AA was shown to be positively correlated with mammary arteriovenous difference of prolactin in lactating sows but not with circulating prolactin concentration (Farmer et al., 2008). Prolactin was shown to increase SNAT2 protein abundance during lactation in rat mammary glands (Velázquez-Villegas et al., 2015) and to regulate LAT1 expression and activity in mammary epithelial cells of dairy cows (Zhou et al., 2020). Thus, prolactin may mediate in part its effect on milk protein synthesis through the upregulation of specific AA transporter proteins. In this study, mammary mRNA abundance of genes encoding transport proteins mediating the uptake of solely cationic AA (CAT-1 and CAT-2b) and the shared uptake of cationic and large neutral AA (ATB^0,+) were not impacted by Arg supplementation and greater prolactin concentrations. It is possible that prolactin regulates these genes at the translation level as shown for SNAT2 and LAT1. Other genes are involved in transport of cationic AA and are well characterized in porcine mammary tissue. They include genes coding for y⁺LAT1/4F2hc, y⁺LAT2/4F2hc, and b^0,+AT1/rBAT proteins (Manjarín et al., 2014). Thus, prolactin could be involved in regulating transport of cationic AA via those genes.

Conclusion

Arginine supplementation tended to increase prolactin concentrations in early lactation without change in mammary transcript abundance of genes encoding for cationic AA transport proteins CAT1, CAT2B, and ATB^0,+, indicating that prolactin may not regulate transport of cationic AA via transcription of SLC7A1, SLC7A2, and SLC6A14. We acknowledge that the number of sows did not allow for assessment of Arg supplementation on litter performance and milk production, therefore caution should be used in interpretation of the data. Nonetheless, current findings show a potential beneficial effect of dietary supplementation with 0.10 g of Arg/kg BW to multiparous sows from day 108 of gestation onward on BW of their piglets on day 1 and throughout lactation. This tendency was likely related to higher BW at birth rather than milk production because ADG was unaffected. Thus, it is possible that Arg impacts piglet birthweight by improving placental angiogenesis. The observed tendency for increased prolactin concentrations in ARG vs. control sows in early (day 4) but not in later (day 18) lactation, as well as the decrease in prolactin concentrations between 2 and 6 hr after feeding are novel findings demonstrating that an interaction exists between prolactin and nutrient intake.

Glossary

Abbreviations

AA

amino acid

BW

body weight

CP

crude protein

DM

dry matter

DNA

deoxyribonucleic acid

MRPL39

mitochondrial ribosomal protein L39

N

nitrogen

NE

net energy

PCR

polymerase chain reaction

PRLR

prolactin receptor

RSP23

ribosomal protein 23

VAPB

vesicle-associated membrane protein-associated protein

Conflict of interest statement

The authors declare no real or perceived conflicts of interest.